IFPRI Discussion Paper 00926 

November 2009 

“Crossing the River While Feeling the Rocks” 

Incremental Land Reform and its Impact on Rural Welfare in China 

 

John W. Bruce 
Zongmin Li 

 

2020 Vision Initiative 

 

 
This paper has been prepared for the project on 

Millions Fed: Proven Successes in Agricultural Development 
(www.ifpri.org/millionsfed) 

http://www.ifpri.org/millionsfed�


INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE 
The International Food Policy Research Institute (IFPRI) was established in 1975. IFPRI is one of 15 agricultural 
research centers that receive principal funding from governments, private foundations, and international and regional 
organizations, most of which are members of the Consultative Group on International Agricultural Research 
(CGIAR). 

FINANCIAL CONTRIBUTORS AND PARTNERS 
IFPRI’s research, capacity strengthening, and communications work is made possible by its financial contributors 
and partners. IFPRI receives its principal funding from governments, private foundations, and international and 
regional organizations, most of which are members of the Consultative Group on International Agricultural 
Research (CGIAR). IFPRI gratefully acknowledges the generous unrestricted funding from Australia, Canada, 
China, Finland, France, Germany, India, Ireland, Italy, Japan, Netherlands, Norway, South Africa, Sweden, 
Switzerland, United Kingdom, United States, and World Bank. 

MILLIONS FED 
“Millions Fed: Proven Successes in Agricultural Development” is a project led by IFPRI and its 2020 Vision 
Initiative to identify interventions in agricultural development that have substantially reduced hunger and poverty; to 
document evidence about where, when, and why these interventions succeeded; to learn about the key drivers and 
factors underlying success; and to share lessons to help inform better policy and investment decisions in the future. 

A total of 20 case studies are included in this project, each one based on a synthesis of the peer-reviewed 
literature, along with other relevant knowledge, that documents an intervention’s impact on hunger and malnutrition 
and the pathways to food security. All these studies were in turn peer reviewed by both the Millions Fed project and 
IFPRI’s independent Publications Review Committee.   

AUTHORS 
John W. Bruce, Land and Development Solutions International, Inc. 
President 
Email: jwbruce@ladsiinc.com  
 
Zongmin Li, Land and Development Solutions International, Inc. 
Vice President 
Email: zli@ladsiinc.com  

Notices 
1 Effective January 2007, the Discussion Paper series within each division and the Director General’s Office of IFPRI 
were merged into one IFPRI–wide Discussion Paper series. The new series begins with number 00689, reflecting the 
prior publication of 688 discussion papers within the dispersed series. The earlier series are available on IFPRI’s 
website at www.ifpri.org/pubs/otherpubs.htm#dp. 

Copyright 2009 International Food Policy Research Institute. All rights reserved. Sections of this document may be reproduced for 
noncommercial and not-for-profit purposes without the express written permission of, but with acknowledgment to, the International 
Food Policy Research Institute. For permission to republish, contact ifpri-copyright@cgiar.org.

mailto:jwbruce@ladsiinc.com�
mailto:zli@ladsiinc.com�
mailto:ifpri-copyright@cgiar.org�


 

iii 

Contents 

Abstract v 

1. Introduction 1 

2. Land Reform in China 3 

3. Land Reform Impacts 18 

4. Asymmetric Tenure Reform and Its Impacts 34 

5. Key Features of the Chinese Reform Experience 39 

References 45 

 



 

iv 

List of Tables 

Table 1. Summary statistics of changes in rural China, 1978 to 1984 4 

Table 2. Township and village enterprises, 1978–94 11 

Table 3. Indicators of rural development, 1978–84 (in 100,000 tons) 19 

Table 4. Per capita income of rural and urban households 21 

Table 5. Comparison of inequality, 1978–86 23 

Table 6. Distribution of land and income in rural China, 1988 and 1995 24 

Table 7. Per capita availability of basic foodstuffs in China, 1949–84 24 

Table 8. Per capita calorific intake, 1965–80 26 

Table 9. Per capita annual consumption of key foodstuffs (kg) 26 

Table 10. Consumption indicators, 1978 and 1988 27 

 
List of Figures 

Figure 1. State and market prices of grain, 1952–90 9 
 

List of Boxes 

Box 1. Follow the Eggs 25 
 



 

v 

ABSTRACT 

Between 1978 and 1984, a massive shift from collective to household agricultural production took place 
in China. These incremental reforms, which Deng Xiaoping called “crossing the river while feeling the 
rocks,” eventually gave 95 percent—160 million rural Chinese families—the right to oversee household 
plots, leading to stunning gains in productivity.1

Despite the success of the HRS, the enhancement of property rights is an ongoing reform process. 
Landholders depended on tenure agreements that could be changed at any time. Rural areas did not have 
the same right to profit from appreciating land values as urban landholders. As cities have expanded 
rapidly, municipalities have requisitioned rural land and issued it to new users at urban prices much 
higher than that paid to the rural villages. The policy debate about the appropriate pace for strengthening 
rural land use rights continues.  

 This reform, the Household Responsibility System 
(HRS), provided strong incentives for farmers to increase labor and improve land, since they could profit 
from any marketable surplus they produced. Meanwhile, the state set quotas and purchased crops, 
providing reliable markets for increased production. It also strongly supported farmers by managing 
irrigation and the agricultural extension system. The state’s earlier investments in rural nonfarm 
infrastructure paid off under the reforms, as workers released from agriculture by the more efficient use of 
labor found employment in local rural industries. In the years following the property reforms, the quality 
of life in rural China improved dramatically: per capita rural income more than doubled from 1978 to 
1984. 

Having examined the substance, process, and effects of the reforms, this paper asks what lessons 
from the reforms are relevant for other developing countries. In spite of differences among countries, 
some elements of the Chinese reform experience seem highly relevant to others engaged in the struggle to 
develop.  

 
 

Keywords: Millions Fed, Food Security, China, Land Reform, Land Tenure, Collectivized Agriculture, 
Household Responsibility

                                                      
1 The Deng Xiaoping quote, suggesting the tentative nature of reform steps, is cited in Gulati et al. 2005, 47. 



 

 



 

1 

1. INTRODUCTION 

In 1977, faced with near famine conditions, poor villages in Anhui Province experimented with a return to 
household farming, which led to impressive productivity gains. Facing up to the weak performance of 
collective agriculture and near famine conditions in many areas of the country, the Communist Party in 
1978 reluctantly embraced these remarkably heterodox experiments. The new system was known as the 
Household Responsibility System (HRS) because parcels of collective land allocated to farm households 
to manage came with obligations to produce specific quotas of key economic crops and to sell them to the 
state at fixed prices. Once permitted, the system spread throughout China like wildfire, so that four years 
later the communes’ land, over 90 percent of the country’s farmland, had been parceled out to more than 
160 million farm households. This remarkably rapid implementation was only possible because the 
reform was largely self-implemented by local communities.  

The state procurement system provided the new family farmers with a guaranteed market for a 
quota of a key economic crop, albeit at below-market prices. After 1977, the state raised its procurement 
prices; in 1979, alone state procurement prices for major crops, on average, increased 22.1 percent. The 
new family farmers lacked full security of tenure; they had only annual land use contracts and had to deal 
with periodic reallocations of land among households to maintain equality in landholding. Nevertheless, 
the opportunity to work the land themselves for the direct benefit of their families elicited a remarkable 
response from farmers.  

By 1983/84, annual household investments in agriculture averaged more than twice the annual 
combined state and collective investments during the same period. Production grew rapidly: crop output 
grew 42.2 percent during 1978–84. The annual growth rates for the three most important crops, grain, 
cotton, and oil-bearing crops, averaged 4.8, 17.7, and 13.8 percent, respectively, between 1978 and 1984. 
(During the preceding 26 years the average growth rates for these crops had been only 2.4, 1.0, and 0.8 
percent per year, respectively.) The changes in farm scale and organization of production, which had 
posed difficulties in many land reforms elsewhere, seem to have been eased by the fact that local 
institutions such as the township and collective continued to support agricultural development. 

While increases in state procurement prices clearly had some impact on production, most studies 
suggest it was surprisingly small: the change in the form of production organization, the HRS reform, 
accounted for between 50 and 85 percent of production increases. Farm families responded to their 
opportunity to sell a part of their production at market prices by investing large amounts of labor and 
inputs to exceed their quota, while at the same time diversifying their production into nonquota crops. 
Labor supervision by the household was far cheaper and more efficient than that of the commune and 
brigade, and with the outflow of surplus labor, labor efficiency and production efficiency generally 
increased. Area cropped area declined slightly but production still grew.  

The labor that had been underemployed under the commune system was now released; it needed 
to find work, and given the tight controls on labor migration, it had to be local work. Townships and 
collectives took advantage of this and of new revenue under the HRS to develop township and village 
enterprises (TVEs), building on the basis of the commune industries and large public investments in 
infrastructure during the commune period. The TVEs were public but produced for the market, and they 
grew rapidly. In 1978, the TVE gross output value was just 7.2 percent of China’s total output value. In 
the 1980s, 60 to 70 percent of rural output value was produced by TVEs. The number of firms increased 
from 1.5 million in 1978 to nearly 25 million in 1994, while the number of employees increased by a 
factor of 4.5 and total output value by a factor of 80. This sector grew far more rapidly than the state 
industrial sector. 

In the years following these reforms, the quality of life in rural China improved substantially, 
reducing (though not nearly eliminating) a long-standing gap between rural and urban households. 
Between 1978 and 1983, per capita rural income more than doubled. The rural poverty rate fell from 76 
percent in 1980 to 23 percent in 1985, and over the 20 years after 1981, it fell to 8 percent. Poverty 
increasingly became a regional issue, related to resource endowments. Food access and affordability 



 

2 

improved. Energy intake increased by 12 percent in the first four years of the HRS reforms (1978–81). 
Energy intake as a proportion of the minimum rural requirement was 122 percent in 1978, and in 1979–81 
it ranged between 140 and 143 percent. After 1978, consumption of foods other than grain rose rapidly, 
with an increase of 40 percent in the areas devoted to nonquota crops.  

By the end of the 1980s, growth rates in agriculture slowed. A policy debate ensued over whether 
it was necessary to strengthen farmers’ rights over their farmland, which remained collectively owned and 
farmed on contract. The government announced gradual increases in the duration of production contracts, 
with longer contracts for areas requiring more investment such as those that reclaimed degraded hillside 
land, and it discouraged frequent redistributions of land. Initially optional, these changes have now been 
built into new land legislation. There is an emerging consensus that China should move toward fuller and 
more marketable property rights in rural land but also a consciousness that land is performing important 
social security and safety net functions. A debate continues on how to balance the benefits to efficiency of 
liberalization of farmers’ rights toward marketability and the losses to social security consequent to 
reforms that would make rights more marketable. Reforms have come painfully slowly.  

Rural households are still seriously disadvantaged. While urban landholders, with more 
marketable use rights, enjoy the benefit of appreciation of the land they hold, rural people do not. This is 
most painfully evident at the interface between rural and urban land, where municipalities compulsorily 
acquire rural land, compensate rural landholders modestly, and then sell that land to urban users at much 
higher prices, generating huge public revenues that have fueled both impressive urban infrastructure 
development and corruption in the use of these off-budget funds. 

The HRS reform and the other reforms it sparked have contributed greatly to the economic 
development of China and the welfare of its citizens. Many developing-country policymakers, in thinking 
through their own reform programs, should consider the lessons learned in China. 



 

3 

2. LAND REFORM IN CHINA 

During the period 1978–84, when Chinese communities implemented a decollectivization reform of 
stunning scope, the management of about 95 percent of farmland in China was returned to more than 160 
million farm households. The gains produced were equally impressive. Production and productivity per 
hectare increased markedly, with readily discernible impacts on the incomes, food security, and 
nutritional levels of Chinese families. The efficiency gains in agriculture released a large amount of 
surplus labor that had existed under collective cultivation, and this in turn fueled the development of 
China’s equally remarkable rural industrialization, accomplished through development of TVEs. In more 
recent years, the government has turned its attention to gradually strengthening the property rights of the 
households that were reform beneficiaries.  

What was the basis in economic theory for this land reform and its success? The economic 
literature recognizes two fundamental institutional reforms in the land sector: land reform and land tenure 
reform. There is a consensus among economists that where labor is relatively cheap, household farming 
enjoys advantages in labor supervision that make it more efficient in using land than larger-scale 
operations, even though the latter may be more highly mechanized. Few larger-scale economies exist, and 
those that do are not in production itself but in primary processing and marketing. Most often, 
historically, large agricultural holdings have not been the product of market forces and have less to do 
with efficiency than with power, domination, and ideology. Land reform is the scaling-down of those 
large units, usually through a process of subdivision, to smaller, more efficient farm sizes. When these are 
given to families in secure tenure, there is a significant distribution of wealth and economic opportunity in 
the society. The principle is the same regardless of whether one is reforming Latin American latifundia or 
collective farms in socialist contexts (Binswanger, Deininger, and Feder 1995).  

The economic literature also asserts that the quality of land rights affects the landholder’s 
investment incentives. Tenure security, the confident expectation of being able to continue in possession 
of one’s land long enough to recoup investments in land, and even to leave the land to one’s heirs, 
strengthens substantially the farmer’s incentive to invest. This is especially important to investments in 
activities such as soil building, tree planting, or terracing, which enhance the long-term productivity of the 
land. The rights to transfer and mortgage land are again asserted to affect both investment incentives 
(since the investment is reflected in the market value of the land) and credit access (because the land can 
be used to secure loans). Land tenure reform involves increasing the robustness of the property rights of 
the producer, increasing incentives to invest and thus the productivity of the land. This is a core 
understanding, reflected in policy prescriptions of the international development community and in 
particular the World Bank (Feder and Feeney 1992; Deininger 2003).  

The literature on rural reform in China typically breaks the rural reform into the “incentive 
reforms” that dominated the period 1978–84 and the gradual market liberalization that began in 1985 and 
extended through the 1990s (see, for example, Huang, Otsuka, and Rozelle 2007). This paper instead 
focuses on the two land reforms. The first reform, the return to household farming from collective 
agriculture, was together with price reforms part of the “incentive reforms” that took place during the 
period 1978–84. The second, a gradual enhancement of property rights, paralleled the market liberation 
beginning around 1989 and ongoing today. The first is considered a spectacular success, while the second 
remains incomplete and its effects are still being assessed.  

The constant throughout these two reforms is what Chinese documents refer to as the “two-tier 
system:” public (collective or state) ownership of land and private-use rights to land rather than full 
private ownership for users. This conceptualization, with its retention of public ownership of land, has 
made it possible for a communist state to embrace a reform process that has moved China ever closer to 
de facto private property in land.  

This paper focuses primarily on the first reform, known as the Household Responsibility System 
(HRS) Reform It explores its impacts on production, investment, food availability, and welfare, but it also 
covers the ongoing process of strengthening producer property rights, with some limited evidence 



 

4 

regarding the actual and potential impacts of this second reform. It finally turns to the “unfinished 
business” of land reform in China and the extent to which lessons learned there may be applicable 
elsewhere. Table 1 summarizes the changes that took place during the reform period.  

Table 1. Summary statistics of changes in rural China, 1978 to 1984  

 1978 1984 %  
Rural households impacted by the land reform (million) 160   
Changes in grain yield (kg/mu)a 169 239* 41.42 
Changes in food production (100,000 tons)b     
Grain 3,047.5 4,070 33.55 
Cotton 21.7 62.5 188.02 
Oil-bearing 52.2 118.5 127.01 
Tea 2.7 4.1 51.85 
Sugar 211.2 396.6 87.48 
Meat (pork, beef, lamb) 85.6 152.5 78.15 
Changes in per capita food consumption (kg)c    
Grain 195.46 249.65 27.72 
Edible oils 1.6 4.66 191.25 
Pork 7.67 12.93 68.58 
Beef and lamb 0.75 1.23 64 
Portray 0.44 1.35 206.82 
Egg 1.97 3.88 96.95 
Aquatic products 3.5 4.32 23.43 
Changes in rural per capita income (% and yuan)d 220 522 137.27 
Changes in calories consumed (% and kcal/day)e 2,226.9 2,450** 45.5 
Changes in rural poverty rate (%)f 33% 11% -22 
Changes in rural nonagricultural labor force (10,000)G 3149.5 5169.6 47.04 
*This figure is for 1988. 
** This figure is for 1980. 
Sources:  
a. Chen, Chen, and Yang, 1992, 536. 
b. State Statistical Bureau, various years.  
c. Zweig 1997, 14.  
d. State Statistical Bureau, various years. Cited in Fan, Zhang, and Zhang 2004, 397. 
e. Wakashiro 1989, 18. 
f. Fan, Zhang, and Zhang 2004, 396. 
g. Chen, Chen, and Yang, 1992, 603. 

Before the Reform: Collective Agriculture 
After the Chinese Communist Party came to power in 1949, it launched a brief program providing land 
parcels to households for them to manage and cultivate. That program was followed by the creation of 
production cooperatives, beginning in 1956, followed in 1958 by a dramatic scale-up to communes, 
averaging 5,000 households and consisting of several production brigades, further divided into production 
teams of 20–30 households. The commune might include as much as 10,000 acres. Work on private plots 
was prohibited (Fan and Pardey 1995, 8). Rural land was “owned” jointly by the commune, the 
production brigade, and the production team, with the production team serving as the basic accounting 



 

5 

unit (Zhang, Li, and Shao 2006, 609). The smallest unit of organization of productive activities was the 
production team, but most tasks were managed at a higher level, such as the brigade. During this period, 
the production team had no real power, and it was not uncommon for the team’s land to be appropriated 
for the collective enterprises of the commune or brigade (Ho 2003). 

The collectivization of agriculture was expected to benefit from the economies of scale predicted 
by Marx and also to provide a base for the development of rural industries. It was also intended, suggests 
Lin (2003, 14), “to serve industrialization by affecting a mandatory drain on rural surplus. Taxation was 
carried out through the control of basic rural production factors, the monopoly in the circulation of 
agricultural products and especially price scissors between industrial and agricultural products.” He notes 
that Chinese economists have assessed the total value drained from agriculture through scissor pricing 
from the 1950s to the 1980s at 600–800 billion yuan. 

The impacts on production of collectivization were disappointing. Grain production in China rose 
during the 1949–52 land-to-the-tiller period by 13.14 percent and continued to increase, if less strongly, 
during the agricultural cooperatives period (1952–58). But with the Great Leap Forward to communes in 
1959, grain production declined. The country suffered serious famine in 1960–63. Grain production 
recovered by 1968 and continued to increase thereafter, though the amount of land cultivated did not 
return to 1958 levels. During the 20 years from 1957 to 1978, the annual income per peasant increased 
from Y87.6 to Y133.6, but the amount of commercial grain contributed by each rural resident declined 
from 85.05 kilograms to 62.6. Shortages of food supplies spread throughout the country, and coupon food 
rationing was introduced in urban areas. Rural areas also suffered from food shortages from the early 
1960s to 1987. Zhang, Li, and Shao (2006, 614-615) estimate that during the 1970s, one-third of the rural 
population did not have a stable food supply. Du Runsheng, the Party’s Director of Rural Policy from 
1978 recently summed up the situation: “Per capita grain production never averaged much more than 300 
kilograms. Of the 800 million peasants, 250 million were impoverished. The nation as a whole could not 
achieve self-sufficiency in grain and required massive imports” (Du 2006, 2).  

Collective agriculture ultimately proved incapable of fostering agricultural productivity 
comparable to that of the “Asian tigers” such as Taiwan and South Korea. Distorted incentives were at the 
base of the failure. In spite of the system’s highly egalitarian aspirations, a work point (gongfen) system 
was introduced in an attempt to create incentives for harder work, but supervision of labor in the brigade 
proved extremely difficult. Ultimately each worker received the same number of work points each day 
regardless of the quality or quantity of his labor. With little incentive for individuals to invest their labor, 
productivity of labor and productivity per land unit declined. Shared tasks without accountability resulted 
in extensive disguised underemployment (Lin 1988, S200; Lin 1990; Lin 2003, 141). Government tried to 
deal with these problems by progressively shifting farm management tasks to lower levels within the 
commune hierarchy. By 1962, the production team had become the basic unit of operation and 
accounting. Decisions regarding farm operations, including the adoption of new technologies, were 
primarily made by team leaders (Fan and Pardey 1995, 8).  

The Reform Begins: Desperation, Experimentation, Replication  
China began to abandon collective production, returning to household farming in the late 1970s. The 
reform process began with local experimentation in Fengyang County in poverty-ridden Anhui Province, 
a poor region plagued by flood and famine, driven by famine and the collapse of confidence in collective 
agriculture.2

HoH 

  

The shape taken by the reform was influenced by farmers’ memories of household farming, even after 
two decades of collective agriculture, and their ongoing experience with their small household food plots. 
Even in the 1960s and 1970s, when the agricultural sector was collectivized, farmers still managed these 
                                                      

2 See Yang and Su (1998) for an account of the politics of famine and reform in China.  



 

6 

small plots. They accounted for about 6.2 percent of cultivated land (Li, Rozelle, and Huang 2000, 5). 
The high productivity of these small “food plots” (usually smaller than 0.02 hectares but several times 
more productive than the collective’s land), suggested the potential of a return to small household farms 
(Zhang, Li, and Shao 2006, 611, 616). In addition, the idea of a return to family farms was not new. Early 
experimentation with bao gan dao hu (contracting everything to the household) went back to the late 
1950s. It spread in Henan Province in 1959 and in southwest and northwest areas of China in 1964. These 
experiments were publicized widely, but were phased out forcefully by government shortly after their 
introduction (Zhang 1998).3

The reform began in 1978 when a few production brigades in Anhui Province secretly distributed 
their land to their member households to farm. The year’s productivity increases were impressive. Some 
brigades in Anhui that returned to household farming had production increases two-to-five times larger 
than those in unconverted brigades (Lin 1988, S221, note 8). Local officials embraced the reform, which 
was then carried out under the protection of Wan Li, the provincial governor of Anhui (Ho 2005, 11).  

 

By 1976, Chairman Mao had died and the Cultural Revolution had come to an end. The country was in 
chaos, and there were grain failures and famine in parts of China. The time was right for the Party to 
reconsider its options. As late as 1977, a return to household farming was specifically forbidden by the 
Central Committee of the Party. Because the performance of collective agriculture had been better in the 
more fertile coastal areas, opposition to reform was considerable in those provinces. Gradually, however, 
the Party came to accept the idea of contracting land to households. Wan Li was promoted from Anhui to 
the Politburo, and the Fengyang experiments received support from influential leaders such as Chen Yun 
and Hu Yaobang. In 1978, the Plenum of the 11th Central Committee of the Communist Party allowed the 
option of breaking up the communal lands into household holdings (Ho 2005, 11). 

At first the Party considered the new system appropriate only for the poor areas of western and 
central China affected by famine, not the more developed coastal regions. Many considered it a temporary 
expedient. The formula under which the change was permitted did not refer to “land reform,” but rather 
conferred “five rights” to production teams: the right to select production crops appropriate to local 
conditions, the right to make rational decisions for management and administration, the right to adopt 
other measures to increase output, the right to distribute products and cash according to their wishes, and 
the right to reject improper instructions from administrative agents (Zhang, Li, and Shao 2006, 611).  

Sustaining the return to household farming against ideological opposition required skillful 
manipulation of ideological themes in the service of pragmatism. Du Runsheng was a key figure in this 
process.4

It is also a general rule that large-scale production – whether capitalist or socialist 
– is superior to the small-farmer economy… But the scale of production cannot be 
determined just by looking at land size. Lenin once said: “A reduction in farm acreage 
during the intensification of agriculture very often implies an increase in the scale of 
production and not a decrease.” He also said, “The principal line in the development of 
capitalist agriculture is changing the economy that is still small-scale in terms of land 

 Du (2006, 3) describes three key points in the reformers’ strategy for winning acceptance of the 
reform within the Party: (1) build the system initially within the communes, rather than abolishing the 
communes; (2) allow a number of forms, among which the populace could choose; and (3) allow the 
reform to spread gradually. In a 1982 speech, Du (1995, 50) provided an ideological basis from the works 
of Lenin for moving to a smaller scale of production:  

                                                      
3 Zhang et al. (2006: 611, 616) provides a detailed account of the return to family farming in late 1978 in the village of 

Xiaogang in Fengyang Country of Anhui Province. 
4After the Third Plenum of the 11th Central Committee of the Chinese Communist Party (1978), Du Runsheng held the post 

of Director, Rural Policy of the CCP Central Committee, and Director of the Rural Department, Research Center for Rural 
Development. He provided leadership in the drafting of the rural-related policy documents for the Central Committee and the 
State Council, including the decisive “No. 1 Documents” on rural development policy, issued each year for five years (1982–86) 
by the CCP Central Committee. These documents, which reflect a remarkable blend of ideology and practicality, allowed the 
HRS reform to keep moving forward before it had fully proven itself and while it was still politically vulnerable.  



 

7 

area into one that is large in terms of scale of production, development of animal 
husbandry, amount of fertilizer use and degree of mechanization”. Clearly the scale of 
production is not equivalent to the amount of land under cultivation. What matters is the 
degree of the organic composition of capital5

The new system spread by voluntary decision of the local collectives.

 and the level of farmland intensity. 

6 It spread rapidly and not 
only in the poorer areas (Ho 2005). In January 1980 only 1.02 percent of all production teams in China 
had changed over to household farming, but by December 1980 the figure was 14.4  percent, then 28.2 
percent by July 1981, and 45.1 percent by October 1981. By 1981, when government finally recognized 
that the HRS reform was broadly applicable, 45 percent of the production teams in China had already 
been dismantled (Lin 1988, S201; Lin 1992, 36; Lin 2003, 144).7

There are few land reforms that can compare with this for rapid implementation of land 
redistribution. In a few years, collective land was contracted out to more than 160 million households (Ho 
2005, 11). Self-implementation by local communities delivered a reform that was incremental and gradual 
but remarkably rapid.  

 By the end of 1983, about 97.7 percent 
of production teams and 94.2 percent of the farm households in China were farming under the new system 
(Lin 1988, S222, note 10). The shift was often rapid. Household farming began to appear in Hunan at the 
end of 1981 and by the end of 1982, 90 percent of the production teams had adopted the system (Lin 
1991, 364).  

The Household Responsibility System Reform 
The new system of contracting land to the household was known as the Household Responsibility System 
(HRS). Land continued to be “owned” by the old production brigade under the collective system, but 
these units began the transition from production organizations to communities.8 A community’s rural land 
under HRS typically existed in one of four categories: residential land, construction land, responsibility 
farmland, and household food plots.9

                                                      
5 The term used by Marx for the ratio of “fixed capital,” or nonlabor costs of production to total capital, including both fixed 

capital and labor costs.  

 The reform was focused on responsibility farmland, the collective 
farmland from the commune period. The reform went through a number of stages in some communities, 
usually within a few years. Lin (2003, 142–143) describes the stages: (1) an initial phase known as the 
group work-contract phase, in which the production team was assigned a quota and the whole group was 
then awarded or punished according to their performance; (2) a second phase called the household output-
quota contract phase, in which a specific output quota and plot were assigned to the household, and output 
exceeding the quota would be given to the household or shared between the household and the production 
team; and (3) the household responsibility system, identical to the second phase except that there was no 
unified allocation of income by the production team. Not all communities went through all three stages, 
and many who did moved through them in successive years.  

6 Lin (1987) analyzes the reform as farmers’ institutional choice, using the model of induced institutional innovation 
developed by Hayami and Ruttan (1985). 

7 For a discussion of the politics and ideological debates around these changes, see Zweig (1997, 62–67). 
8 Ho (2001) analyses what he describes as the “deliberate institutional ambiguity” involved in the statement that the rural 

collective owns the land. The “administrative village” (the successor to the brigade) most often did the contracting of land to 
farm households, and it was often not clear whether it was this administrative village or the “natural village” (the successor to the 
production team) that owned the land. Inconsistent use of terms such as “farmers’ collective,” “collective economic 
organization,” and “villagers’ committees” in instructions of land confounded cadres responsible for implementing them.  

9 Rural homes were privately owned by households, though the land on which they stood was still owned by the collective. 
Housing sites were obtained without charge by residents and could not be transacted, though they could be inherited with a house 
if the heir was a resident of the collective. The house plot and the household food plot have been the most secure tenure niches in 
rural China, with no term limit. Construction land was land allocated to public facilities and enterprises based on a contract for 
building. In this case there was considerable variability of terms among the contracts within and among communities, depending 
on the enterprise and the use.  



 

8 

Under the HRS, the collective farmland was distributed to households to cultivate as “responsibility 
land,”10

Tenure to users of farmland was provided through contracts from the brigade’s executive 
committee. These were one-to-three-year contracts with households and included an obligation on the part 
of farmers to produce specific amounts of staple crops (rice, wheat, and some others) for sale to the state 
at fixed prices (hence the emphasis on “responsibility” in the term Household Responsibility System). 
Crops produced over these quotas could be sold on the open market, as could crops not covered by 
quotas. The use contracts could involve relatively modest use charges, could be inherited by resident 
heirs, and could be subleased but not otherwise transferred. One finds variations among communities, 
because the contracts were produced locally (there was no national model contract), but this is the broad 
pattern (Wang 2005).  

 usually with some services (tractor plowing, irrigation maintenance) being provided by the 
production teams. It was distributed strictly according to household size, without reference to the size of 
the labor force. The distribution was highly egalitarian, and as a result the household’s holding on the 
average was fragmented into an average of nine tracts, even though the size of a holding was only about 
1.2 acre (Lin 1989, 16). Regulations prohibited sales of use rights, rentals, and use of land as collateral of 
both privately and collectively managed land (Li, Rozelle, and Huang 2000, 6).  

The highly egalitarian ethos of the villages led to frequent redistribution of the responsibility land 
(referred to as “adjustments”) to ensure equality of holdings as families changed over time. 
Redistributions were often at the initiative of the collective but also took place to accommodate 
government projects that needed land. Other factors were in play as well: a desire on the part of 
government to prevent development of too strong a sense of land proprietorship on the part of farmers, 
and on the part of collective cadres, a desire to take advantage of the considerable rent-seeking 
opportunities such redistributions offered (Wang 2005; Rozelle et al. 2005). Land redistribution 
frequencies varied considerably. Li, Rozelle, and Huang (2000, 6) found that more than 90 percent of the 
villages in Hubei had readjusted in the past 15 years, but in Sichuan, the percentage with readjustments 
was only 22 percent.  

While security of tenure left something to be desired under the early HRS, the HRS still had 
manifest advantages. Ling (1991, 148) summarizes them: the flexibility to use family resources to deal 
with temporary labor shortages, the ability of the family to adjust expenditures, the low management costs 
of self-management, the production unit being conterminous with the consumption unit, and the fact that 
the income and living standards of households now depended on themselves and how they managed their 
farm. The farmers in Ling’s study pursued their comparative advantage, reducing areas under grain and 
shifting to oil-bearing seeds, tobacco, and vegetables and moving into animal husbandry (Ling 1991, 
106).  

The new household farmers had one important advantage over many land reform beneficiaries, 
which often found their input supply and marketing chains badly disrupted or still controlled by former 
landlords. The beneficiaries of the HRS reform had a guaranteed market for a quota amount of their major 
crops at state-set prices and the opportunity to sell their over-quota production and nonquota products to 
the state at market prices.  

A Parallel Reform: The State Procurement System  
The sad state of Chinese agriculture in 1977 led government to introduce a significant reform of the 
system of state procurement of agricultural products and procurement prices at the same time that the 

                                                      
10 Li, Rozelle, and Huang (2000) indicate that at the outset of the reform, local leaders allocated the collective land to 

peasants in three tenure types: ration land (kouliang tian), to meet household subsistence requirements, responsibility land (zeren 
tian), given on the condition that farmers deliver a low-price grain or cotton quota to the state, and contract land (chengbao tian), 
auctioned off by village leaders for a fee. Not all villages had all three categories of land, and land tenure types differ sharply 
among the villages: while all villages had responsibility land, a minority had contract land and less than 20 percent had ration 
land.  



 

9 

HRS reform was getting underway. This was in fact the primary government response to the food crisis. 
The HRS was a more spontaneous development later embraced by government. 

In 1977, government was the only legal purchaser of many key commodities, including rice, 
wheat, maize, oilseeds, and cotton. Provinces were assigned quotas by central government and these were 
broken down geographically, so that production teams, at the base of the pyramid within the communes, 
were assigned quotas. Quota production had to be marketed to the state at prices set by the state. With the 
division of communes into household farms, these quotas became the responsibility of households, 
assigned by the same contract that allocated them their farmland. The state was in these years effectively 
a monopoly purchaser of quota production, of grain and key agricultural products produced in excess of 
quotas, and often even of nonquota crop production. 

After 1977, the state procurement prices were raised to increase farmers’ incentives. In 1979 
alone, state procurement prices for major crops, on average, were increased 22.1 percent. The government 
also began to increase grain imports and loosened restrictions on private interregional trade in agricultural 
products (Lin 1992, 34, 39). Government purchasing prices rose greatly in the years after 1979, those for 
grain by 100 percent and those for many other crops by 40–50 percent, with free market prices still higher 
(Ling 1991, 106). Huang (1998, 57) provides procurement prices for 1952–90 (Figure 1).  

Figure 1. State and market prices of grain, 1952–90 

 
Source: Huang 1998, 57. 

At the same time, the state introduced important reforms in the state procurement system. From 
1977 onward, farmers were allowed to trade grain on free markets once they fulfilled their delivery quotas 
to the state procurement system. Bans that prohibited peasants from growing cash crops were terminated. 
Farmers regained the right to grow vegetables or other nonquota cash crops and to sell their products in 
the open markets. Trade in urban areas was still restricted in the initial stage, however, and most markets 
were still substantially controlled by the government (Jia and Fock 2007, 2).  

Since 1988, a series of reforms have sought to further change the incentive structures of grain 
production, consumption, and exchange. To enhance farmers’ incentives to undertake grain production 
and exchange, either the state purchase price has been raised or the quantity purchased at the lower state 
price has been reduced, depending on the measures adopted in the marketing reforms (Kung 1992, 165–
166). To secure low consumption prices in favor of urban areas and industry, the domestic grain market 
was rationed. The state raised grain prices a number of times after 1979. Grain rationing was abolished in 
the early 1990s, but the government remained intent on keeping grain prices low and supply stable for 
urban areas.  



 

10 

Control on retail prices in urban areas was retained, keeping urban food cheap (Ling 1991, 112). 
Consequently, as pricing of agricultural production was liberalized, the fiscal burden on government rose 
substantially. Subsidies by the state for food consumption by the urban population increased substantially 
to Y27.5 billion per year in 1985, or a fourth of the wage bill of all state employees (Lin 1989). Du (2006, 
9–10) attributes the problems in the years after 1986 to government increasing the grain purchase price 
without correspondingly raising the price at which it would be sold to city people: “Thus the more grain 
production increased, the greater the financial subsidy, and massive increases of grain bought at higher 
prices created a burden too heavy for the state finances to bear…”  

Grain prices fell sharply during the good harvests, such as occurred in 1991 and 1992, and after a 
glut of grain in 1983/84, government in 1985 cancelled the policy of paying 50 percent more for the grain 
procured beyond the contract amount and instead purchased all grain at an increased average price (Oi 
1999, 619). In the years after 1985, the major variable in determining incomes became marketing, as 
government switched back and forth on key policies in an attempt to balance rural incomes and urban 
food prices. There was an attempt in the early 1990s to raise state procurement prices to market level and 
to dismantle the parastatal marketing system (Jia and Fock 2007, 2). However, food prices rose sharply 
and the state compulsory quota system was again re-imposed for most parts of China in 1995 but at a 
lower procurement level (Huang, Bi, and Rozelle 2004).  

The manner in which this procurement and market reforms were managed is one of the distinctive 
features of the Chinese rural reforms of the 1980s. Whereas most post-Communist countries have gone 
the “big bang” route, with a sudden transition to market prices for agricultural production, China opted for 
a two-track approach, maintaining quotas and set prices for quota production while liberalizing markets 
for nonquota production and allowing markets to control prices for above-quota production of quota 
crops. Chinese and other commentators have suggested that the dual-track approach achieved efficiency 
without creating losers, continuing to enforce the existing plan while simultaneously liberalizing the 
market as a method of making implicit lump-sum transfers to compensate potential losers of the reform 
(Lau, Qian, and Roland 2000).  

To Lin (2003, 332–333), it is this “crossing the chasm” between state prices and the market “in 
two steps” that constitutes the genius of China’s agricultural pricing reforms. The approach, he urges, 
provided the state with an assurance of sustained grain production and farmers with an assurance of a 
predictable if modest farm income during a period of great uncertainty. At the same time, it provided 
strong incentives for farmers to (1) exceed quotas and so to be able to market above-quota production 
freely, and (2) to diversify into nonquota crops. Zhang (1998, 165) emphasizes the role that 
experimentation played in the reform and what he calls “institutional learning” through experimentation. 
Regarding the pricing reforms, he notes that under the two-tier price system “… government was able to 
control certain agricultural products, ensuring stability of supply, while some products were allowed to 
enter the markets. This allowed market mechanisms to develop and consumers and producers to adapt 
gradually to a market environment.”  

A Parallel Development: Rural Industrialization  
With the transition to household farming, it became evident that substantial disguised surplus labor had 
existed in collective agriculture. In 1985, the main foodgrain crops required fewer labor days per hectare 
than in 1978—down 22 percent for rice, 48 percent for corn, and 53 percent for wheat (Vermeer 1989, 
85). The labor freed up had to find other local employment. Those thrown off by agriculture could not 
migrate elsewhere to work because of the hukou permit system, a remarkably effective system of strict 
residence control, which kept labor in their existing rural and urban areas (Zhang, Li, and Shao 2006, 
620).  

The solution to the problem of rural labor surplus was rural industrialization, reflected in the 
slogan “Leave the farm but not the village, move to the factory but not the city.” The rapid development 
of the agricultural sector under the HRS reforms created not only a labor surplus but also a surplus of 
funds for local enterprise development (Lin 2003, 311). Township and village governments inherited the 



 

11 

collective factories upon the break-up of the communes, and they seized on the opportunity presented by 
cheap, local labor. Rural industrialization was only later embraced by government as policy (Pei 2005). 
The TVEs became the success story of Chinese development in the 1990s. As with the HRS, the TVEs 
were not the brainchild of planners but a spontaneous reaction to market opportunities by local 
governments. They were public but autonomous, and their production and other decisions were market-
driven.  

The rapid launch of rural industries was facilitated by continued collective ownership of land by 
the successors to the communes (Pei 2005). In 1978, the TVE gross output value was just 7.2 percent of 
China’s total output value. In the 1980s, rural industry became the leading force behind rural economic 
development, with 60–70 percent of rural output value produced by township and village enterprises (Lin 
2003, 147, 311). Zweig (1997, 258) shows the dramatic growth, with the number of firms increasing from 
1.5 million in 1978 to 17.5 million in 1987 to nearly 25 million in 1994, while the number of employees 
increased by a factor of 4.5 and total output value by a factor of 80. Table 2 provides more detailed data 
on the development of TVEs between 1978 and 1994. By 1996, China’s rural enterprises employed more 
than 135 million people, about one-third of the rural labor force. These laborers came from the labor 
released by the de-collectivization of agriculture, and yet agricultural production continued to grow (Oi 
1999, 620).  

Table 2. Township and village enterprises, 1978–94 

Year  No. of firms No. of employees Total output 

 (1,000s) (1,000s) (100 million RME) 

1978 1,524.2 28,265.6 493.07  
1979 1,480.4 29,093.4 548.41  
1980 1,424.6 29,996.7 656.9  
1981 1,337.5 29,695.6 745.3  
1982 1,361.7 31,129.1 853.08  
1983 1,346.4 3,2346.4 1,016.83  
1984 6,065.2 5,2081.1 1,709.89  
1985 12,224.5 69,790.3 2,728.4  
1986 15,153.0 79,371.4 3,540.9  
1987 17,502.4 88,051.8 4,764.3  
1988 18,881.6 95,454.5 6,495.7  
1989 18,686.3 93,667.8 7,428.4  
1990 18,504.0 92,647.5 8,461.6  
1991 19,078.8 96,091.1 11,621.7  
1992 20,792.0 105,811.0 17,975.4  
1993 24,529.0 123,453.0 31,540.7  
1994 24,945.0 120,175.0 42,588.5  
Sources: Zweig 1997, 258. Number of firms and employees, and the total output value come from the State Statistical Bureau 
1995.  

During the 1980s and early 1990s, this expansion of employment in the TVEs was much more 
rapid than that in the state industrial sector. Some of the TVEs were poorly located and, as markets 
became more open, a process of consolidation took place. By the mid-1990s, local governments had 
begun privatizing the TVEs. With the support of the center, they shifted preferential treatment to private 
firms. By the end of 1997, about a third of all collectively owned rural enterprises had been privatized, the 
largest number turned into shareholding companies. Local authorities could afford to shift their support 



 

12 

because private firms were finally becoming a viable alternative source of tax revenue (Zhang 1999; Oi 
1999, 624).  

The impact of the HRS, which was extended through the TVEs post-1985, may prove to be more 
important than the direct effects felt during 1978-84. Lin (2003, 312) notes that TVE development, 
sparked by the HRS reform, has driven subsequent reforms: “TVE development was a market driven 
process, and greatly pushed the economic reform toward a market economy, corrected the distorted 
industrial structure, and became part of a dual track system in areas of both resource allocation and price 
formation which placed the traditional system under growing pressure.”  

Rural industrialization and the new prosperity did, however, have a downside for the agricultural 
sector. Land in cultivation declined in the post-reform years, especially in the coastal provinces where 
fertile soils are located. According to Du, Tang, and Zhang (2002), about 30 percent of the diminished 
cultivated land in those areas was occupied by collective-owned industries and 20 percent was used for 
peasants’ dwellings.  

The Property Rights Reform  
The initial impact of the reforms seems to have played out by the end of the 1980s. Fan, Zhang, and 
Zhang (2004, 397) explain that during 1985–89, rural income continued to increase but at the much 
slower pace of 3 percent per year. Rural residents earned less than half of what their urban cohorts earned 
in 1978; rural income was 42 percent of that in urban areas. Due to the success of rural reforms, that 
percentage increased to 59 percent in 1984. However, it declined again to 36 percent in 2000, owing 
mostly to fast growth in urban areas, state manipulation of rural–urban terms of trade, and relatively 
sluggish increases in rural earnings. Rising inequality of rural incomes was apparent after 1995, with 
absolute incomes even falling at the bottom end of the income distribution. Benjamin, Brandt, and Giles 
(2005) attribute most of this decline in welfare to lower agricultural incomes brought about by lower farm 
prices (cited in Fan, Zhang, and Zhang 2004, 25–27). Agricultural production ceased to grow so rapidly 
and urban incomes grew faster than those in rural areas, leading to an ever-widening gap between urban 
and rural incomes.  

The impacts of the HRS had been so remarkable that the ideological lobby for recollectivization 
had become muted, but old ideas about economies of scale (big is good) persisted. Many Chinese experts, 
including Zhang (1998), attributed lack of greater investment in agriculture to the egalitarian land 
distribution, arguing that it prevented the development of larger farms which could take advantage of 
economies of scale. In the late 1980s, the continuing outflow of labor from agriculture, most pronounced 
in the coastal provinces, engendered a number of experiments with administratively managed “scaling-
up,” that permitted some large operating units to allow for more effective deployment of mechanization. 
The scale experiments involved larger farms managed on contract by private farmers, village 
cooperatives, or other enterprise forms. Most reviews of their performance suggest that they were highly 
subsidized but not more efficient than smallholder agriculture.11

                                                      
11 Bruce and Harrell (1989) skeptically reviewed scale production experiments in Wuxi, Wu, and Changshu counties of 

Southern Jiangsu Province; in Shunyi County near Beijing; and in Nanhai County of Guangdong Province. The selection of the 
“scale farmers” by the collective often had little to do with managerial ability. Prosterman, Hanstad, and Ping (1998) and 
Prosterman and Bledsoe (2000) also review scale operations critically, noting that the international experience has demonstrated 
that there are few economies of scale in agricultural production (as opposed to primary processing and marketing). They found 
that scale operations in Jiangsu and Zhejiang Provinces were producing at about the same level as smallholders, in spite of 
extensive subsidies, and were equally unconvinced by the scaling-up approach (shareholder systems) practiced in Nanhai.  

 Experimentation with such arrangements 
continues, and spontaneous local solutions make it increasingly difficult to generalize about land tenure 
arrangements in Chinese villages. Zhang (1998) reviews a number of ongoing experiments, including (1) 
the “two-field” system, which allocates a subsistence plot to each household and market production plots 
only to those who wish to farm them and to pay for access to them, and (2) the wasteland contracting 
system, in which those who contract to develop wasteland receive very long-term use rights, which are 
fully transferable and mortgageable (see also Hanstead and Ping 1997). The decentralization of land 



 

13 

administration allows a considerable variety of arrangements within the framework of collective 
ownership.  

Alongside the scale debate, another policy debate has gone on over how and when farmers’ land 
rights should be strengthened. The initial HRS contract was annual, and terms could be extended. At the 
same time, however, the debate focused on periodic reallocations of land among farm households, and 
their disincentive effects on investments in land. Such reallocations could override the terms of contracts 
with farmers. Reallocations involved high transaction costs. During the land redistribution in Wugone 
Country after the summer harvest of 1989, 10,000 labor units (persons) per day were set to measuring and 
dividing farmland for one month, at a cost of roughly Y800,000 (Ling and Jiang 1993, 457). The relative 
merits of market or administrative allocation of land are contested ground among researchers (Carter and 
Yao 2005).  

The actual frequency of such reallocations is also contested. A sample survey from the Research 
Center for Rural Development (RCRD) in 1988 showed that 65 percent of 250 sample villages 
nationwide had redistributed land every two years since 1983. The distribution in 91 percent of those 
villages was caused by demographic changes, while only 5 percent reallocated land in order to enable 
farm size to grow, and less than 2 percent did it for plot consolidation (Ling and Jiang 1993, 447). A 2001 
survey conducted under the auspices of RCRD over 1,600 households in 17 provinces found that four-
fifths of the villages had conducted at least one land readjustment since the first allocation after the HRS 
was instituted. In the great majority of villages that had conducted at least one land readjustment, the most 
recent adjustment had occurred within the past five years (DRC and RDI 2002, 22).  

The Development Research Center (DRC) and Rural Development Institute (RDI) study (2002, 
24–30) found a broad awareness that government favored a reduction in the frequency of such 
reallocations (94.1 percent), and those supporting that policy outnumbered opponents 5:1. However, when 
asked about a complete prohibition of readjustments, a slight majority opposed a complete prohibition. 
Interestingly, those whose villages had never had a readjustment more strongly supported prohibition of 
readjustments. They found that more than a third of the contracts and certificates issued to those farmers 
had provisions allowing land readjustments even during the term of long-term contracts. Asked if 
readjustments would continue, 45.8 percent of respondents expected them to continue, whereas only 12.2 
percent felt they would definitely not continue. Reforms to enhance property rights have also been the 
subject of policy experiments, notably in Meitan Country, Guizhou Province.12

These national policy discussions of scale, reallocations, and property rights have led to further 
reforms, but through a very gradual process of legal change. Now, for the first time, Chinese 
commentators on land reform have begun to use the language of property rights, rather than 
responsibilities, having clearly entered the domain of tenure reform. The process of incremental land law 
reform that has taken place is described briefly below, without doing justice to its complexity. The uneven 
implementation of specific tenure reform initiatives makes it difficult both to generalize about the 
situation on the ground and to assess confidently the tenure reform’s impacts.  

  

At the outset, there was no law regarding the HRS, and reforms came through a series of rapidly 
evolving party pronouncements and instructions. This was not unusual in China during that period. China 
strongly exhibited the Marxist contempt for law as a tool of oppression in the hands of the bourgeois 
state. It conveyed less of a sense that law could also be a tool for change than, for instance, Russian 
communism. After flirting briefly with more formal legal modes early on, Mao in 1959 dramatically 
downgraded the role of law and law professionals in Chinese society (Peerenboom 2003, 43–45). The 
approach to normative innovation was ad hoc experimentation, tweaked through a succession of party 
edicts. A basis in law for the new institution would only be provided once the new model had proved 
itself. In China, a new law was not associated with innovation but was seen as evidence that a new 
arrangement had succeeded and was to be enshrined in law. Du (2006, 5) explains  

                                                      
12 The Meitan experiment, elements of which were later widely replicated in Guizhou Province, involved an end to 

reallocations, strengthened use rights, increased marketability (and even mortgageability) of those rights. It is discussed in Bruce 
and Harrell (1989) and at a later stage by Zhou (1994).  



 

14 

As recognized by institutional economics, forming a stable system must be a process in which the 
populace chooses for itself. This process included different sides in mutual dialogue that leads to 
coordination and integration, according to the requirements of the interest and political pursuits of each 
side. Given that the Party wanted to give the populace a free choice, we did not need to turn this practice 
into a law of the state for the time being. We had to treat law as an outcome of a social choice and 
eventually provide legal guarantees in the form of law.  

The basic land administration arrangements in the wake of the HRS were first set out in law as 
the Land Administration Law of 1987 a decade after the reform began. A number of important reforms in 
the HRS were being urged by central government by this time. Rural Work Document No. 1 of 1984 
urged local officials to extend the duration of farmland contracts to 15 years and for longer periods for 
special uses such as reforestation and fruit tree plantations. The Politburo’s No. 5 Document of January 
1987 (“Deepening the Rural Reforms”) encouraged farmers who had moved into nonagricultural 
employment to transfer their use rights to others. To this end, the Constitution’s prohibition of all 
transactions in rural land (Article 10) was amended by the National People’s Congress on April 12, 1988 
(Bruce and Harrell 1989).  

The provisions of the new 1987 Land Administration Law reaffirm collective ownership of rural 
land (Article 8) and allocation to households for use under contracts (Article 12). It does not include some 
of the tenure-strengthening measures recommended in Party pronouncements, which were permissive 
rather than mandatory. Local cadres could, and often did, ignore them. During the author’s field work in 
Fujian in the mid-1990s, he realized that while some responsibility contracts were being made for longer 
periods, they still contained clauses providing for reallocations at the discretion of the collective, 
nullifying the greater security of tenure provided to farmers by the longer terms (Bruce and Muo 1988).  

Amendments to the original Land Administration Law, the most recent in 1999, provide for 
issuance of certification and registration of land rights of rural collectives and of user rights over rural 
land for nonagricultural purposes. They require registration of changes in ownership and use of land and 
buildings. They also provide for administrative settlement of land disputes, overall land use planning, and 
the establishment of a land survey system and a land statistics system. Most important, the 1999 
amendments require a minimum term of 30 years for farmland. Longer terms, up to 70 years, are 
available for specialized uses such as forestry or agroforestry on hillside lands and for some construction 
projects.  

The provisions of the Land Administration Law on contracting of farmland remained sketchy. In 
2002, after several years of intensive debate in Party circles, a full legal framework for the HRS was 
provided. The Rural Land Contracting Law became effective on March 1, 2003. It does not replace the 
Land Administration Law, but supplements it. The law confirms a number of administrative orders that 
had strengthened the use rights of rural land users. Now land use contracts must be for 30 years for arable 
land, 30–50 years for grassland, and 30–70 years for forest land. They must be in writing and signed by 
both parties. Readjustments are restricted. Land is not to be readjusted (redistributed by the collective) 
during the contact term, except if this is required by a natural disaster and or unspecified “other special 
circumstances.” Any readjustment during the term of the contract must be approved by two-thirds of the 
members of the village assembly or two-thirds of the village representatives. The village may maintain a 
part of its land in a flexible reserve to adjust landholdings for newly added village population, and it may 
also use reclaimed land and land returned voluntarily by contracting parties for this purpose. The 
contacted land use rights are now clearly inheritable during their terms, an important change.  

The township government is responsible for rural land contracting and contract management 
within its administrative jurisdiction and often provides the needed forms and instructions. The contracts 
are effective when concluded but are required to be registered. The county or higher level of government 
is required to issue to a user a land contracting and operation certificate and to register the contract. The 
Ministry of Agriculture keeps these registries. Fees are not to be collected beyond what is necessary to 
cover the cost of the certificate.  

While the collective cannot normally take back the land during the contract term, the rapid 
growth of urbanization and its impacts are recognized. When land users move to a small township, they 



 

15 

can retain their use rights, but if they move to a city and change their household registrations to 
nonagricultural, they must surrender the contacted land. In that case compensation is to be paid for 
investments on the contracted land. The Rural Land Contracting Law confirms the right of the holder to 
transfer (assign), lease, exchange, or otherwise engage in transactions regarding the use right. For 
assignments and exchanges, the permission of the collective must be sought, but for other transactions the 
collective must only be notified. The transaction must be made in writing if it is for more than a year, and 
it must be registered. If it is not registered, the rights are subject to the good faith claims of a third party.  

A collective may contract rural land to a unit or individual outside the collective but only with the 
approval of two-thirds of the village assembly or two-thirds of the village representatives as well as the 
approval of the township government. Provision is made by the Rural Land Contracting Law for 
mediation and arbitration of disputes concerning contracts and for ultimate recourse to the courts, as well 
as for civil and criminal liability for officials and others who violate the procedures required by the Law 
or the rights of land users under the Law. The new law went into effect on March 1, 2003.  

Studies suggest that implementation of the reforms under the 1999 Land Administration Law and 
the 2003 Rural Land Contracting Law has been very uneven (Prosterman, Schwartzwalder, and Jianping 
2000; Prosterman, Ping, and Zhu 2004). They have stressed that while rural land has long remained 
outside the market economy, land use rights in urban land, which is owned by the state rather than the 
collective, has long been readily marketable. Assets in the form of long-term use rights to urban land have 
become an important element in the major wealth differential between urban and rural households. This 
dichotomy introduces major distortions into China’s land economy, a matter discussed in the section 
below on unfinished business. Advocates of stronger rural land rights have continued to assert the need 
for greater security and marketability of rights in farmlands, to allow them to gradually achieve a market-
determined value, thereby eliminating the strong duality that now exists between rights in rural and urban 
land.13

At the close of the Fifth Session of the Tenth National People's Congress on March 16, 2007, the 
Congress approved the new Property Law, which took effect on October 1, 2007. It is more important for 
its content on urban land than on rural land. Regarding collective agricultural land, it largely refers to the 
2003 Rural Land Contracting Law. It does, however, provide an important clarification of the nature of 
collective ownership and some other matters, summarized in Zhu et al. (2007, 12–13) and Development 
Research Center and World Bank (2007, 6–7):  

 In 2007, China launched a comprehensive new Property Law, a first and modest step toward 
unifying rural and urban land tenure. 

• Collective land is owned not by the local collective institutions, but is “collectively owned by 
members of such collectives.” This makes it clear that members have a property interest in 
that ownership, and the language of the Law suggests that compensation should go directly 
into the hands of affected farmers because they are part owners of the property right taken 
away by the state. In the past, compensation went primarily to the collective. 

• Farmers’ 30-year land rights are for the first time characterized as property rights (as 
opposed to their contractual rights). This should in theory at least make it more difficult to 
take away farmers’ land rights.  

• National standards now exist for real property registration. The Law requires that the 
establishment, change, transfer, or termination of real property rights be registered to have 
legal effect, unless otherwise provided by law. Regulations are to be separately promulgated 
to govern the registration process and specify the authorities that will be responsible for 
handling registrations. China now has 250 million farm households, with each household 

                                                      
13 Wang (2005) and Prosterman , Ping, and Zhu (2006) both are excellent and relatively up-to-date summaries of the legal 

and regulatory framework for land, but both were prepared during discussions leading up to the enactment of the new Property 
Law, which was still in draft form at the time, and so are not the final word. Wang (2005, 82), commenting on the draft property 
law, worries that “the transitional process” of land tenure reform will be “frozen in time or halted through legislation.” 



 

16 

having on average five plots, so nationwide land registration and certification will be a 
challenge.14

• Farmers have modestly improved prospects for extended duration of their land rights, as the 
Law provides in Article 126, which says that “contracting farmers should continue extending 
the contract according to relevant law.” This is not as strong as the wording used for urban 
land use rights, which speaks of automatic renewal (Article149), but it is still the first time 
Chinese law has expressed a presumption in this regard in relation to the 30-year term. 
Article 130 affirms the limitations on readjustments in the Rural Land Contracting Law, but 
breaks no new ground in this regard. 

  

• Compensation criteria for rural lands now include a requirement that compensation 
arrangements must protect farmers’ livelihoods. While the specific criteria from early law 
and policy pronouncements remain unchanged from the multipliers of average annual crop 
yields under the Land Management Law, Article 42 of the Property Law specifies that 
“…social security benefits [should] be arranged for affected farmers so that their livelihood 
and rights be protected.” This is a positive development and could open the way for payment 
of compensation in excess of the Land Management Law standards, but the requirement is 
vague and could be easily evaded by officials.  

Reformers were in some ways disappointed by the new Property Law. It does not allow rural land 
rights to be used as collateral for loans. Rather than incorporating important new reinforcement of the 
property rights of rural landowners, it tends to consolidate progress already made.15

The future direction of land law reform in China is suggested by a recent policy document 
referred to as the October 2008 CPC Central Committee Decisions. This is a broad policy document 
aimed at a more balanced and integrated rural-urban development. It sets a goal of doubling rural per 
capita incomes by 2020 and includes several major decisions on rural reform and development, with a 
new land policy at its core. Under the new policy, the “existing land contracting relationship shall remain 
stable and unchanged for a long time.” The precise meaning of this statement remains under debate. In a 
recent magazine interview (Caijun, December 12, 2008), Chen Xiwen (Director of the Central 
Government Leading Group of Rural Affairs Office) urged that “The only option is to implement no 

 Provisions allowing 
mortgaging of rural land appeared in earlier drafts, but they were dropped in the final version. Some 
commentators have noted that mortgaging of rural land is not prohibited. Similarly, the Law does not 
directly address the marketability of rural construction land, including housing plots. Again, such 
transactions are not expressly prohibited. These failures to specify reforms leaves the door open for their 
promulgation by party edict at a later date. And it is worth noting that the structure of the law itself 
provides an indication of the reform direction of the future. Instead of having separate sections on urban 
and rural land, it provides rules by type of use right (farmland, construction land, and so forth) and makes 
distinctions within those categories if considered necessary. This seems to suggest an intention to unify 
the systems in the future.  

                                                      
14 Government had already launched a registration pilot with FAO support nearly two years earlier (Project 

TOP/CPR/30088, “Rural Land Registration and Certification Pilot,” begun in 2005). The project is led by the Central 
Government Leading Group of Rural Affairs Office, with participation by the Ministry of Agriculture, the Ministry of Land and 
Resources, and the State Council’s Legislative Office. The sites selected are two villages in Feidong County of Anhui Province. 
Significant inconsistencies among the land contracts on record are said to have been a major challenge. A draft strategy document 
recommends moving next to further pilots at different locations and in different economic environments, including urban pilots 
(FAO 2009).  

15 A cautious assessment in a Development Research Center and World Bank report (2007, 1) says that “[The] Property Law 
is unmistakably an advance, it is just as clearly part of an evolutionary process. Indeed, as will be shown, the power of the 
Property Law lies not in that it represents a sudden departure from what has come before, but in the fact that it builds upon and 
consolidates successive policy instruments and legal reforms that have taken root over the last three decades. And by the same 
token, the Property Law is not the end of a journey – as this paper will explore, multiple challenges remain to ensure that land 
policy and its implementation fully supports China’s development objectives.”  



 

17 

readjustment in response to change in household size, and cut off the link between household size change 
and change in household landholdings. All household land rights acquired through the Rural Land 
Contracting Law should not be readjusted in the future.” Others in government are less open to a 
perpetual land use right, which they feel would reduce state and collective ownership to a legal 
technicality, much like the residual interest of the Crown in privately owned land in English Law.  

The Decisions go on to state that “…farmers shall be allowed to transfer, lease, exchange, assign 
or join as stock shares land contracting and operation rights legally, voluntarily and in return for adequate 
payment and develop multiple forms of proper scale farming.” This is a strong affirmation of the 
transferability of rural farmland, though mortgaging is still excluded. This seems to suggest that a 
decision has been made to rely on market forces to accomplish any scaling-up of farm sizes that may be 
needed as labor continues to move out of agriculture. 

Finally, and most important, market transactions for commercial purposes in collectively owned 
construction land will now be allowed. This is the most important reform item in this package. It is not 
entirely clear whether it applies to farmers’ residential land, but it is nonetheless a first crack at the state 
monopoly of the market in land for commercial uses. This reform has major implications for land 
requisitioning and compensation for requisitioned land, since this land and the improvements upon it will 
now have a market value. The Decisions say explicitly that the ultimate objective is a unified market in 
construction land, whether it is state or collectively owned. 

The October 2008 Decisions, together with the Property Law passed in March 2007, reflect 
China’s intention to gradually narrow the gap between land tenure in the rural and urban sectors. The 
reforms will eventually require amendments to the Land Administration Law and the Rural Land 
Contracting Law. Implementation is some years away, though in some provinces, such as Guangdong, 
pilots for the sale of collective construction land are already underway with administrative approval. 
China is moving gradually, sector by sector and use by use, into an era of greater transferability of rural 
land use rights. Ultimately, it seems clear to the authors, an integration of urban and rural land markets is 
intended. This would allow China’s rural people to benefit from the appreciating value of their land, as do 
urban landholders.  



 

18 

3. LAND REFORM IMPACTS 

The impacts of the first land reform, the HRS reform, have been carefully studied and analyzed. A 
remarkable degree of consensus exists regarding both the positive impacts of the reform and the major 
role it played in initiating broader rural reforms. Those studies focus on the key years 1978–84. In the 
case of the second land reform, the property rights reforms, the situation is quite different. Control of farm 
households over their land has increased gradually, in many small steps, and implementation has been 
halting. It remains partial. Factions in the central government have differing visions of the appropriate 
rate of implementation, and local officials, who have vested interests in their control over land allocation, 
have often stalled changes. The Chinese court system, while it is improving, is not yet a reliable tool for 
enforcing rights against officials. Most of the econometric studies are thus tentative. However, some 
unanticipated results have emerged that deserve consideration.  

HRS Reform Impacts 
A number of indicators have been used to assess the impact of the HRS reforms. These include 
investment in farms, farm productivity, farm income, and, to a lesser extent, food availability and 
nutrition. Reforms often involve a cocktail of reform measures, and a major challenge in assessing their 
effects is disentangling the relative importance of the results achieved by the different measures. 
Fortunately, there are some excellent studies of impacts for the critical years 1977–84, during and 
immediately after implementation of the reforms, before the waters became muddied by many other 
changes.  

Investment Impacts 

Public investment in agriculture was substantial under the commune system. Was this public investment 
continued after the HRS reforms, and to what extent was it supplemented by private investments of the 
new household farmers?  

From 1979, when the HRS had begun to spread rapidly, the government’s investment in 
agriculture declined from Y6.24 billion, reaching Y3.43 billion in 1983, then partially recovered to Y4.68 
billion in 1987. Meanwhile the collectives’ investment decreased from Y8.71 billion in 1979 to Y2.07 
billion in 1985. These declines were however offset by per household investment of Y34.49 in 1981, 
increasing to Y100.13 in 1983, and remaining at an annual level of Y80–100 thereafter. Between 1983 
and 1984, annual household investments averaged Y18.10 billion, more than twice the annual combined 
state and collective investments during the same period and more than the annual combined state and 
collective investments before the reforms (Feder et al. 1992, 2).  

Cumulatively while farmer investments in agriculture were substantial, they were dwarfed by 
farmer investments in residential construction. Housing investment was 85 percent of farmers’ total 
investment in 1982, remaining high throughout the 1980s and still reaching 70 percent in 1992. Feder et 
al (1992, 13) find that the ratio of housing investment to productive investment in the counties they 
studied ranged from 2.71 to 10.88 during the period 1983/84–1987/88. This rush to invest in housing 
reflects the insecurity of tenure on productive land, which was subject to periodic land reallocations, 
compared with the more secure tenure on residential land. Farmer investments in land were for the most 
part not investments in land itself, but in capital stocks such as farm equipment and livestock (Feder et al. 
1992, 13).16

Was investment by the new HRS landholders credit-constrained? Feder et al (1989, 525) indicate 
that institutional and informal credit hardly existed in the 1978–84 period. It is now common, but their 

   

                                                      
16 Investments in capital stocks (livestock and equipment) more than doubled in the five-year period covered by the study 

(1983/84–1987/88), with average annual capital growth of 15 percent over the period covered. Fewer than 10 percent of 
households owned tractors, but a large number of households invested in livestock for food (pigs and poultry) and draft power 
and even boats in one site with extensive canal networks (Feder et al. 1992, 11).  



 

19 

study found very few instances of the medium- or long-term credit usually required for land 
improvements. Long- and medium-term credit was and is a constraint, they suggest, and they indicate that 
most funds for investment in land are generated by off-farm income of the farm household and relatives.  

Public investment continues to be important. Neglect of facilities was a concern of those who 
analyzed the HRS reforms early on (Bruce and Harrell 1989). Irrigation is the life blood of Chinese 
agriculture, and major irrigation works remained a public responsibility. In any event, the transfer of 
management of these facilities to new institutions seems to have gone more smoothly than might have 
been anticipated. The later literature does not note any significant problems. The ability of government to 
maintain these services to the beneficiaries of the HRS reforms in part accounts for the impressive growth 
and poverty reduction delivered by the reforms.  

Production Impacts 

Research on the effects of the HRS has focused primarily on its impact on productivity, and especially on 
teasing out the relative importance of institutional reforms, pricing reforms, technological change, and 
public investment in achieving the major productivity gains achieved in the years 1978–90.17

Lin (1988, 1989) found in his analysis of provincial-level panel data a 42.2 percent output growth 
in the cropping sector in 1978–84. Between 1978 and 1984, the key years for HRS implementation, the 
annual growth rates for the three most important crops, grain, cotton, and oil-bearing crops, averaged 
respectively 4.8, 17.7, and 13.8 percent. This compared favorably with the average growth rates of 2.4, 
1.0, and 0.8 percent per year, respectively, for those crops in the preceding 26 years, 1952–78. Zhang, Li, 
and Shao (2006) estimate grain area output per unit under the HRS to be nearly 25 percent higher than 
that under collective farming. They also find that national grain output rose about 300 million tons in 
1978 and increased to 407 million tons in 1984. As Zweig (1997, 75) points out, this explains why grain 
output could rise even while the acreage under grain decreased. In Table 3, Zweig provides 1978–84 
production rates of various quota and nonquota crops and other agricultural products: 

  

Table 3. Indicators of rural development, 1978–84 (in 100,000 tons) 

 1978 1979 1980 1981 1982 1983 1984 

Grain output 3,047.5 3,321.2 3,205.6 3,250.2 3,534.2 3,872.8 4,070 

Cotton output 21.7 22.1 27.1 29.7 36.0 46.4 62.5a 

Oil-bearing crops 52.2 64.4 76.9 102.1 118.2 105.5 118.5 

Tea output 2.7 2.8 3.0 3.4 4.0 4.0 4.1 

Sugarcane 211.2 215.1 228.1 296.7 368.8 311.4 396.6 

Meat production  85.6 106.2 120.6 126.1 135.1 140.2 152.5 

(pork, beef, lamb)        

Sources: 1978-81: Zweig 1997, 75; 1982: State Statistical Bureau 1982; 1983: Beijing Review, no. 35 (August 27, 1984); and 
State Statistical Bureau Communique, April 29, 1984; 1984: USDA 1985. 
a. Ministry of Agriculture 1985. 

During 1984–88, grain output declined only slightly, but per unit area of output improved. The 
national grain output increased to 505 million tons in 1996 and reached a high of 512 million tons in 
1998. In Henan, where the HRS was implemented in early 1983, per mu yields of all three key products 

                                                      
17 The analysis here looks at the contracted responsibility land as a whole, but there were important subcategories, in 

particular hillside land. This land, often neglected and denuded in the collective period, has been developed aggressively under 
agroforestry and tree farming, since the advent of the HRS. Initial allocations of forest land to households on the same model as 
farmland unfortunately resulted in deforestation through unsustainable cutting. Later, the emphasis shifted to re-afforestation 
under long-term contracts, with an emphasis on agroforestry. This has been far more successful and is an important 
environmental success of the HRS reform. See Bruce, Rudrappa, and Li (1995) and Zhang and Kant (2005).  



 

20 

rose suddenly in that year, by 25.3 percent for grain, 89.3 percent for cotton, and 35.7 percent for oil-
bearing seeds (Ling 1991, 104).  

Early studies ask to what extent these impressive output increases were due to the institutional 
reforms under the HRS and how much they owe to the state price increases during the same period. Lin 
(1992, 46–47) finds that output growth during 1978–84 was 42.2 percent. He concludes that 45.8 percent 
of this output growth came from increases in inputs, the most important being increases in fertilizer 
applications, which alone accounted for about one-third of the output growth. The HRS reform accounted 
for 48.6 percent of output growth, as much as the combined effects of the various input increases. 
Changes in market prices and state procurement prices did not affect productivity, but they did contribute 
nearly16.0 percent of output growth, probably through input use, cropping intension, or crop mix.  
This is roughly in line with the two other serious econometric analyses. MacMillan, Whalley, and Jing 
(1989) found that of the total farm productivity increase in 1978–84, 41 percent of the increase in the 
cropping and animal husbandry sector could be attributed to total factor productivity growth; of that, 78 
percent was attributable to the farm institutional reform (HRS) and 22 percent to price increases. Wen 
(1989) found that farm output increased by 56 percent due to the institution of the household-based 
farming system. Zhang, Li, and Shao (2006) reach similar conclusions.  

However, Fan (1991) raises an interesting question about the relative contribution of 
technological change to the production growth in Chinese agriculture. Huang and Rozelle (1996) explore 
this issue with respect to rice. They suggest that technology adoption (adoption of hybrids and the move 
to double cropping) was the most important determinant of yield growth during 1978–84, accounting for 
nearly 40 percent of the change. They find that 35.6 percent of the rice yield improvements were due to 
the HRS (Huang and Rozelle 1996, 362). They acknowledge that institutional innovation may have 
contributed more to the growth in agricultural output generally, the topic of the studies by Lin and others 
noted above.18

Fan, Zhang, and Zhang (2004, 408) raises another important question: the role played by decades 
of public investment in the success of the growth after the rural reforms. They suggest that the economic 
reforms that began in the late 1970s could not have achieved such rapid economic growth and poverty 
reduction had there not been several prior decades of massive government investments in rural 
infrastructure, especially in irrigation, from 1953–76.

 Huang, Otsuka, and Rozelle (2007, 19) note that Fan (1991) and Huang and Rozelle 
(1996) conclude that accounting for technological change, institutional change during the late 1970s and 
early 1980s contributed about 30 percent of output growth.  

19

Sorting out the relative role of the various measures is difficult because they interact and re-
enforce one another. But there is a broad consensus built on solid survey data and sophisticated 
econometric analysis that the institutional reform creating the HRS was the primary factor responsible for 
remarkable output growth between 1978 and 1984, accounting for roughly 40–60 percent of that growth.  

 Investments in infrastructure and research and 
development were slow in the 1980s, but major investments in education after 1978 resulted in a 
substantial improvement in labor quality. Fan, Zhang, and Zhang argue that public investment played a 
significant role in the growth of agriculture and poverty reduction during 1978–84, accounting for 12 
percent of growth and 45 percent of poverty reduction. They concur, however, that 60 percent of 
production growth in Chinese agriculture between 1978 and 1984 was due to the rural reforms.  

                                                      
18 Lin (1991) reviews adoption of hybrid rice in Hunan Province and the impact of the introduction of the HRS on adoption. 

He finds that during the collective period, adoption of hybrids did not correspond to economic rationality and was the result of 
political promotion. After the transition, economic rationality (farmer costs and benefits) became the major factor in adoption. 
Adoption of hybrids continued to rise, although the reforms slowed it by disrupting the network consisting of the county, brigade, 
and team research and extension.  

19 Gulati, Fan, and Dafali (2005, 8) note that after 1953 top priority in public investment was assigned to irrigation, which 
grew at an impressive rate in the pre-reform period. Irrigated area as a percentage of arable land grew from 23.3 percent in 1953 
to 26.2 percent in 1957, an increase of 5 million hectares during the first five-year plan period. Ling (1991, 106) indicates that up 
to the end of 1983, in the administrative district of Laouyang Prefecture in Henan Province alone, government invested Y 450 
million, to complete 1 billion cubic meters of earth and stone work. A total of 40,000 projects for flood control, drainage, water 
and soil conservation, electric power works, and irrigation were carried out.  



 

21 

These impressive levels of agricultural output growth could not be maintained. While overall 
agricultural growth rates averaged 7.9 percent during 1978–84, they dropped to 4.1 percent during 1984–
87, still better than the 2.9 percent for 1952–78 but a clear decline (Lin 1992, 35). The decline after 1984 
is attributed by Lin (1992, 87) to the drop in the rate of growth for fertilizer use from 8.9 percent during 
1978–84 to 3.7 percent during 1984–87. There was also a swift outflow of labor from the cropping sector. 
The growth rate of the agricultural labor force dropped to 2.9 percent in 1984–87, compared with 8.6 
percent during 1978–84. These declines were due in part to a sharp reduction in state procurement prices. 
After 1985, although agriculture still grew at the respectable rate of 4.1 percent, the boom in crop 
production came to an end; the outputs of grain and cotton declined with decreased amounts of land under 
cultivation. Because the Chinese leadership had long been “grain fundamentalists” who saw self-
sufficiency in grain as a central national security concern, some questioned the reform process and even 
led to discussions among the leadership of possible re-collectivization; at the same time others used these 
decreases in production to argue that the reform needed to be carried to its conclusion by full privatization 
(Lin 1992, 3).  

Income and Poverty Impacts  

Per capita income increased to Y 522 in 1984 from Y 220 in 1978, a growth of 15 percent per year in real 
income per capita at 1990 prices, contrasting sharply with the pace of the pre-reform period of 2.3 percent 
per year (Fan, Zhang, and Zhang 2004, 396; Gulati, Fan, and Dafali 2005, 15). Lin (2003, 311), deducting 
for price factors, estimates that productive net income per capita increased from Y 166.39 to Y 291.10, 
while per capita cash-in-hand plus the outstanding amount of the savings deposit in rural areas (year-end) 
increased from Y 26.6 to Y 85.3. Between 1978 and 1983, per capita rural income more than doubled, 
rising from Y133.6 in 1978 to Y 310 in 1983, significantly improving in relation to urban incomes (Table 
4, taken from Renwei 1993,82).  

Table 4. Per capita income of rural and urban households 

 Income per  capita 
(yuan) 

Year  Rural Urban Ratio of urban to rural incomes 

1957 73 254 3.48 

1964 102 243 2.38 

1978 134 316 2.36 

1979 160 377 2.36 

1980 191 439 2.30 

1981 223 500 2.24 

1982 270 535 1.98 

1983 310 573 1.85 

1984 355 660 1.86 

1985 398 749 1.88 

1986 424 910 2.15 

1987 463 1,012 2.19 

1988 545 1,192 2.19 

1989 602 1,388 2.31 

1990 630 1,523 2.42 

Sources: Renwei 1993, 82; Ling 1991; State Statistical Bureau various years; State Statistical Bureau 1984b. 



 

22 

Some household income continued to come from collective activities, but this decreased radically 
over the same period. In 1978, 66.3 percent of family income derived from the agricultural collective and 
only 26.8 percent from family production, while in 1989, 82.2 percent of family income came from 
family commodity production and only 9.4 percent from the collective sector (Zong 1993, 282).  

Ling (1991, 155) notes that the rate of growth in peasant average per capita income (at constant 
prices) still increased by 2.1, 9.8, and 7.1 percent in 1985, 1986, and 1987, respectively. Ling makes an 
important point: those increases were due not just to increases in farm output and rises in prices of farm 
products, but to an increase in the number of income sources, notably from employment in rural industries 
or housing construction. Still, even in 1998, agriculture accounted for 60 percent of rural household 
income (Oi 1999, 622).  

How did these increases in income affect poverty levels in China? In the 20-year period after 
1981, the proportion of the population living below the poverty line fell from 53 percent to 8 percent. Half 
the decline in poverty came in the first few years of the 1980s. In 1980, a staggering 98 percent of China’s 
poor lived in the rural areas, and the bulk of the dramatic reduction in poverty during the first years of that 
decade came from rural areas. The rural poverty rate as calculated by the Chinese government fell from 
76 percent in 1980 to 23 percent in 1985. Between 1980 and 1985, relative inequality between rural mean 
income and urban mean income fell to less than half its 1980 level, flattening out in 1985–90, and rising 
again to 1980 levels during in the 1990s. Agricultural growth did more to reduce poverty and inequality 
than either the secondary or tertiary sectors. The share of the urban population in poverty rose from 19 
percent in 1980 to 39 percent in 2002 (Ravallion and Chen 2004, 1–7, 52). 

A World Bank publication (Harrold 1992, 20) gives somewhat different figures, but it also points 
to a dramatic decline in poverty during the relevant period. It indicates that the proportion of the 
population below the poverty line declined from 22 percent in 1978 to less than 10 percent by the middle 
of the 1980s. By 1990, some 11.5 percent of the rural population and 0.4 percent of the urban population 
remained in poverty, but still, more than 160 million people had emerged from poverty during the reform 
era.  

How were the increases in income distributed? One of the issues raised concerning the HRS 
reforms has been the extent to which they have worsened inequality in incomes. Working off such a 
radically egalitarian and relatively uniform livelihood base as that in rural China, it seems likely that the 
reforms would have increased inequality, and researchers ask to what degree this was the case. In fact, the 
distribution of income improved by most measures during the early part of the reform period, as average 
incomes rose substantially with only a modest increase in inequality. The Gini coefficient of income 
distribution in the rural areas declined steadily from 0.32 in 1978 to 0.22 in 1982, but thereafter inequality 
began to increase and the Gini coefficient rose to 0.34 in 1988 and on to 0.42 in 1995 (Zong 1993,78, 
Griffen, Khan, and Ickowitz 2002) (see Table 5).  
  



 

23 

Table 5. Comparison of inequality, 1978–86 

Two World Bank estimates of rural Gini ratios 

 (1) (2)  
1978 … 0.32  
1979 0.257 0.28  
1980 0.237 0.26  
1981 0.231 0.23  
1982 0.225 0.22  
1983 … 0.25  
1984 … 0.27  
1985 … 0.30  
1986 … 0.31  
Sources: Khan et al. 1993, 61; Column 1: World Bank 1985, 29-30; Column 2: Calculations made by a World Bank working 
group on Poverty in Developing Counties, quoted in Ahmad and Wang, 1991, 46.  

It appears that the disparities in incomes that developed after 1982 were due to more efficient 
utilization of resources or differences in production capacity or the amount of physical inputs (Ling 1991, 
110, 124). Following the introduction of market mechanisms, regional income disparities among farm 
households did grow and social tensions were acerbated, but the inequality was not very pronounced and 
was more equal than the distribution of nonfarm income (Ling 1991, 147, 154). Still, jealousy of wealthy 
villages and extortion by cadres increased (Zong 1993, 78).  

Griffen, Khan, and Ickowitz (2002, 312) ask whether the rise in income inequality after 1988 was 
one of the consequences of restoring a household farming system. Working with land holdings adjusted 
or unadjusted for irrigation, they recalculate the Gini coefficient for 1988 and 1995. They find that the 
distribution of land became more equal between 1988 and 1995. It was not the HRS reform, they 
conclude, that was the source of growing inequality in rural incomes after 1988, but the growing 
inequality in nonfarm sources of rural incomes (Table 6.) Fan, Zhang, and Zhang (2004, 397) attribute 
this worsening inequality to a growing differential in rural nonfarm opportunities among regions. Writing 
more recently, Khan and Riskin (2005) and Yingying, Hua, and Harrel (2008) reach the same conclusion: 
the greatest factor contributing to the development of regional inequality in rural incomes was rural 
enterprise. In earlier stages, wages from rural industry were the key factor, but by 1995 ownership of rural 
small businesses emerged as the greatest source of rural income inequality. Assets, they note, have 
replaced labor as the source of income inequality.20

  
  

                                                      
2