Journal of Rural Studies 43 (2016) 145e158

Contents lists available at ScienceDirect

Journal of Rural Studies

journal homepage: www.elsevier.com/locate/jrurstud

Characteristics and driving factors of rural livelihood transition in the

east coastal region of China: A case study of suburban Shanghai
Zhaoxu Liu, Liming Liu*
College of Resources and Environment, China Agricultural University, Yuanmingyuanxi Road, Beijing 100193, China

a r t i c l e i n f o a b s t r a c t

Article history: Based on data from 412 households, this study used the sustainable livelihoods approach to analyse
Received 29 November 2014 changes in livelihood strategies and the relationship between those changes and rural household atti-
Received in revised form tudes in an eastern coastal region of China. Focussing on household off-farm employment decisions and
16 October 2015
land transfer decisions, we classified the households into six types based on their livelihood strategies,
Accepted 12 December 2015
namely, off-farm employment-oriented, balanced, off-farm, idle, professional, and conventional. The
Available online xxx
survey results indicate that two capital factors, age and education, and two mediating factors, wage and
crop prices, are important influences of off-farm employment decisions, whereas factors such as policies
Keywords:
Livelihood strategy
and subsidies, machinery, activities organized by local agricultural cooperatives, and land rent sub-
Livelihood transition stantially influence land transfer decisions. By analysing the characteristics, trends and driving factors of
Rural household livelihood transitions, i.e., changes in rural household livelihood strategies, from 1993 to 2013, we
China explored the mechanisms of livelihood transitions. The results indicate that the proportion of conven-
tional households decreased and the proportion of off-farm employment-oriented households and off-
farm households increased. Agricultural production transformed from a system of scattered house-
holds (conventional and off-farm employment-oriented households) into broad-scale management
households (professional and balanced households). These changes were the main characteristics of
livelihood transitions under rapid industrialization combined with rapid urbanization in the Qingpu
district. Rural livelihood transitions exhibited a stepped pattern, going through an initial stage, a self-
organizing stage, and a government-led stage.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction Accompanied by this restructuring process that was driven by

Since reform and opening-up was launched in 1978, China's
traditional central planning economy has gradually transformed
into a market economy (Long et al., 2012; Siciliano, 2012). In eastern
coastal China, the processes of urbanization and industrialization
have resulted in profound changes in the vast rural areas (Liu et al.,
2011; Long and Woods, 2011). These changes affected agricultural
restructuring, structural employment transformations, large-scale
farming and technology innovations (Liu et al., 2014; Long and
Woods, 2011; Siciliano, 2012). During these processes, agricultural
production exhibited diversified operation modes and functions
and shifted from quantity orientation to quality and benefits
orientation (Liu, 2014; Long et al., 2010; Mullan et al., 2011).
* Corresponding author. College of Resources and Environment, China Agricul-
tural University, No. 2 Yuanmingyuanxi Road, Beijing, 100193, China.
E-mail address: liulm@cau.edu.cn (L. Liu).
related policies, were tremendous livelihood strategy changes
among rural households.
Recently, the changes in rural livelihoods in China have attracted
the interest of academics involved in a variety of research, such as
quantification and evaluation of livelihood assets (Chen et al.,
2013a; Herrmann and Fox, 2014; Li et al., 2014), livelihood
vulnerability and risk analysis (Fang, 2013; Guo and Zhou, 2013;
Liao et al., 2014), influencing factors and diversification of liveli-
hood strategies (Fang et al., 2014; Hogarth and Belcher, 2013;
Hogarth et al., 2013; Huber et al., 2014; Lu and Lora-Wainwright,
2014; Tang et al., 2013), impacts of policies and institutions on
livelihoods (Chen et al., 2013b; Galipeau et al., 2013; Groom and
Palmer, 2012; Li et al., 2013; Zhen et al., 2014) and the in-
terrelationships between livelihoods and the environment (Kemp
et al., 2013; Nath and Inoue, 2014; Urgenson et al., 2014; Zheng
et al., 2013). Despite considerable progress, most insights into ru-
ral livelihood changes are based only on snapshots. However,

http://dx.doi.org/10.1016/j.jrurstud.2015.12.008
0743-0167/© 2015 Elsevier Ltd. All rights reserved.
146 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158
livelihoods are usually dynamic (Ellis, 2000; Scoones, 2009), and as
such, they involve transitions between different strategies
(Mushongah and Scoones, 2012). Therefore, many variables
involved in the process of rural livelihood transitions must be
researched, including on- and off-farm employment, income
sources, resources bases, financial capacity and social relationships
of the households (Conway and Chambers, 1992; Scoones, 1998). To
close this gap, Ulrich et al. (2012) presented a bi-temporal inter-
and intra-household comparison, which covered 13 years, on the
portfolio composition of assets in Kenya. Mushongah and Scoones
(2012) used a combination of a bi-temporal household surveys,
in-depth biographical interviews, and wealth rankings to examine
livelihood changes in southern Zimbabwe over a 20-year period.
Radel et al. (2010) comparatively assessed the agricultural liveli-
hood transition and the accompanying land changes in Mexico's
southern Yucatan region. These studies employed a comparative bi-
or multi-temporal perspective to assess changes in livelihood as-
sets or strategies of rural households. However, these comparative
survey-based analyses provide limited insights into the continu-
ously changing processes of livelihood transition. As livelihood
transition is a continuous process, livelihood system changes are
best investigated and analysed from a dynamic process perspective.
However, the lack of long-term data at such detailed levels impedes
the understanding of livelihood dynamics.
In light of recent calls for greater efforts to understand liveli-
hood dynamics (Addison et al., 2009; Scoones, 2009), this paper
contributes to the current livelihood literature with a continuous
process analysis of the rural livelihood transition over 20 years. In
this paper, a survey-based exploration of rural livelihood transition,
a case study in the east coastal region of China, is presented.
Because statistical data at the household level were not available,
this study aims to explore livelihood changes using a household
survey. We focus on the typical livelihood assets (labour and land)
and the accompanying typical livelihood strategies (off-farm
employment and land transfers), which are identified by the
important role they play in China's rural livelihoods. By investi-
gating and backtracking the current and former decisions that
households made regarding labour and land, we develop a typology
of livelihood strategies to define each of the household's current
and former livelihood types, by which the footprint of each
household's livelihood strategy and the continuously changing
processes of livelihood transition can be restored. Accordingly, this
paper contributes to the literature on the impact of personal per-
ceptions and attitude factors on livelihood decisions. Household
decision making is influenced by a combination of internal factors
of the household and external factors from the environment and
from the socio-economic context (Mattison and Norris, 2005;
Valbuena et al., 2010). Internal factors include personal percep-
tions and attitudes towards making a certain decision (Yu et al.,
2013). For example, even rural households with the same socio-
economic characteristics may have different perceptions and atti-
tudes toward their livelihoods, which may cause them to make
different decisions. However, few studies have examined the
empirical evidence of households' attitudes towards their liveli-
hood decisions (Jongeneel et al., 2008; Meyfroidt, 2013). To close
this gap, this study surveyed and analysed the household attitudes
toward multiple factors that influence off-farm employment and
land transfer decisions.
The objectives of this paper are (1) to identify and classify the
main livelihood strategies of rural households in the face of socio-
economic transitions in eastern coastal China, (2) to survey and
analyse farmers' attitudes toward multiple factors that influence
livelihood decisions regarding off-farm employment and land
transfers and (3) to explore continuous livelihood changes of the
households and how long-term shifts in livelihood strategies
emerge. Section 2 of this paper describes the study area and
context, while Section 3 describes the concepts and methodology of
the study. Section 4 then describes the results of the rural livelihood
transition survey, presents statistical relationships between farmer
attitudes and behaviours, and investigates evolutionary character-
istics, trends and driving factors of rural livelihood transitions.
Sections 5 and 6 discuss and conclude the main findings and
indicate the potential for further application of the results.
2. Study area and context
2.1. Study area
The case study area is situated in Qingpu district, which is
located in western Shanghai. Shanghai (31.1
N, 121.4
E), situated on
an alluvial plain at the mouth of the Changjiang River (Fig. 1), is the
largest city and the most economically prosperous region of China.
As the economic, financial, trade and shipping centre on China's
east coast, Shanghai occupies an area of approximately 6400 km2
and has a population of over 24 million (NBSC, 2012). In 2012, the
land area of Shanghai comprised 0.06% of China and the population
was less than 1.8%, but its gross domestic product (GDP) accounted
for 3.8% of China's overall GDP (NBSC, 2012). Shanghai has expe-
rienced rapid economic development since the beginning of the
1990s. From 1990 to 2012, its resident population grew from 13.34
million to 23.80 million, the GDP increased from 781.66 billion
RMB¥ to 20,181.72 billion RMB¥, and the shares of primary industry,
secondary industry and tertiary industry in the GDP changed from
4.4%, 64.7%, and 30.9%e0.6%, 37.2%, and 62.2%, respectively (SSB,
2010). Rapid urbanization and industrialization have had pro-
found influences on the development of the suburban rural areas.
Although agriculture in Shanghai has an extremely low production
value share, its grain output per unit, per capita net income of
farmers and commodity rate of agricultural product all rank near
the top in China. Its typical role in rural development has some
reference value for the agricultural development of many regions in
the context of rapid urbanization and industrialization.
Qingpu has a resident population of 1.16 million, of which 29.2%
is registered in the rural area (SSB, 2010). In 2012, the shares of
primary industry, secondary industry and tertiary industry in the
GDP were 1.4%, 57.5%, and 41.1%, respectively (SSB, 2010). In the
agricultural sector, plantation and aquaculture activities contribute
to the production of the main fraction of the gross agricultural
output value. The main cultivations are paddy, wheat, vegetables
and fruit. Among these, vegetables, paddy, and wheat represent the
most productive cultivations with respect to income generation. It
is a typical zone with a long history of rice-farming. Dianshan Lake,
located in this area, is the most important drinking water source for
Shanghai. Located in the water source protection area of the
Huangpu River, this region implemented strict surface water
quality standards, which led to strict constraints for large polluting
industries, such as heavy industry, livestock and poultry farming. As
a pilot region for agricultural reform in China, agricultural pro-
duction and operations in the Qingpu district involve a higher de-
gree of industrialization and standardization. The phenomenon and
problem that appeared in the livelihood transition process will
have important significance for similar areas that are beginning to
or are already experiencing livelihood transitions.
2.2. Policy and institutional context
To better understand the process of rural livelihood transition in
Qingpu, we focused specific attention on China's socio-economic
transition and the relevant policies and institutional context for
the past two decades. Because China's market-oriented economic
 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158
Fig. 1. Location of the case study area.
reforms were expedited at the end of 1992, the rapid development
of the manufacturing and service sectors resulted in an increased
demand for cheap labour. Meanwhile, the development of modern
technology resulted in improvements in agricultural productivity,
creating a large amount of surplus labour. Hence, ruraleurban
migration became a significant feature of China's economic devel-
opment beginning in the mid-1990s (Chang et al., 2011). However,
most of the migrations represented a temporary and individual
pattern due to China's unique labour and land systems. The
household registration system (HRS), which was established by the
central government in 1951 and 1958, was initially established to
prohibit rural labour migration. While this migration restriction has
been gradually relaxed, it still limits migrant labourers from
enjoying the same institutional, economic, and social statuses as
local residents (Whalley and Zhang, 2007). The method adopted by
rural migrants to respond to this institutionalized discrimination is
to keep the land of their original home as a safety net for when the
get old or in the event they lose their job. On the other hand, land
tenure arrangements characterized by small farm size and frequent
reallocation due to the egalitarian distribution principle, make it
risky to migrate for a long period because the migrants may lose
their land in the next session of land reallocation (Yan et al., 2014).
China has a distinct land tenure system. The household re-
sponsibility system (HRS), which was adopted by China in the early
1980s, replaced collective farming and separated land-use rights
from land ownership for the first time (Dong, 1996; Hu, 1997;
Watson, 1983). Through fixed-term contracts, village collectives,
known as the village-based governing body, officially own the land,
but individual households gain the rights to use it for their own
production activities (Yu et al., 2013). Initially, the land was
distributed mainly according to household size, and the contract
period was only 15 years. Because of egalitarian principles, each
household received several plots allocated based on soil fertility,
irrigation conditions, location, etc. (Yan et al., 2014). Small, highly
fragmented parcels, and frequent adjustments thus became the
salient characteristics of household farmland in China. The average
farm size of a Chinese household was approximately 0.5 ha that
147
was fragmented into several plots. Additionally, the reallocation of
land depended on demographic change and threatened the tenure
security of rural landholders.
Facing the above-mentioned challenges, China began to build
relevant systems and establish policies to promote the efficient use
of both land and labour, which were vital to China's industrializa-
tion, urbanization, and rural development. On the one hand, the
central government put in place a series of measures to improve the
comparative benefits of agriculture. For example, it abolished local
fees levied on rural households between 2002 and 2003 and
embarked on the grain subsidy programme in 2004. On the other
hand, the central government introduced a series of legal measures
to intensify land tenure security. Key among these were the 1998
revision of the Land Management Law (Chen and Davis, 1998) and
the 2003 Rural Land Contracting Law (RLCL) (Deininger et al., 2014).
Over time, the land use rights have become closer in nature to
private property rights (Mullan et al., 2011). While the ownership of
land is officially in the hands of the rural collectives, the contracted
farmlands can be autonomously used for agricultural production by
households to meet family needs, and the rights to use farmlands
can be transferred to other households for agricultural activities
through rent, exchange or inheritance (Yu et al., 2013). In this study,
the term “contracted land” refers to land that is allocated to indi-
vidual households from village cooperatives, and “transferred land”
refers to land that has been rented to households. Land exchange
and inheritance are not included in this study as they do not involve
decisions that households make to change their livelihood strate-
gies. As the most economically prosperous region in China,
Shanghai faced problems regarding labour and land much earlier
than the rest of China, and thus, Shanghai was the nation's pioneer
with respect to experimenting with institutional innovations. In
addition to the national institutional arrangements, from 2006,
Shanghai has pursued the moderate scale management policy to
promote land transfers and the efficient use of farmland, which
focuses on the normalization of the form and content for trans-
ferring land and on the providing of cash subsidies to moderate
scale tenant farmers according to the acreage of land they rent.
148 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158
3. Concepts and methodology
3.1. Conceptual framework
The concept of livelihood has been defined in a variety of ways
by various authors. The most common definition defines livelihood
as ‘a means of gaining a living’ (Chambers and Conway, 1992). In the
last few decades, several frameworks have been developed by
various institutions (e.g., FAO, DFID, UNDP, etc.) to analyse the
sustainability of livelihoods. The DFID's (the UK Department for
International Development) conceptual framework focuses on the
conceptual organization of the factors that contribute to liveli-
hoods, such as five capitals, vulnerability of livelihoods, institu-
tional process and organizational structure, livelihood strategies,
and outcomes (DFID, 1999). Livelihood capitals refers to the asset
base of the community and of different categories of households
(FAO, 2005). In addition to the conventional assets, such as financial
(e.g., income, credit, liquid assets), natural (e.g., land, forest, water),
and physical capital (e.g., buildings, machines), household assets
include various elements of human capital (e.g., skills, knowledge)
and social capital (e.g., networks, relationships). Livelihood strate-
gies is the range and combination of activities and choices that
people make (i.e., ways of combining and using assets) to achieve
their livelihood goals (DFID, 1999; FAO, 2005). This framework has
some unique attributes as it offers a checklist of important issues
for development and poverty research, and it summarizes the
connections between these issues. Furthermore, it reminds people
to focus on the critical effects and processes, and it emphasizes the
multiplicity of interactions between various factors affecting rural
household livelihoods.
Livelihood transition is essentially the changes in livelihood
strategies in an area at a given time (Bhandari, 2013; Mushongah
and Scoones, 2012) that are driven by socio-economic change and
innovation. The conception of livelihood transition may be further
developed as follows: livelihood transition refers to the changes,
which are driven by socio-economic change and innovation, in
livelihood strategies in a certain region over a specified period of
time, and it usually corresponds to the transition of asocio-
economic development phase.
The sustainable livelihoods approach informed our data collec-
tion and analysis in several ways. First, according to the concept of
livelihood strategies as defined in the SLA, our study of rural live-
lihood transition focuses on the combined activities and choices in
which rural households engaged when using their livelihood as-
sets, especially the changes in these activities and choices. Second,
the classification system built to distinguish the livelihood strate-
gies of rural households is also based on the concept of livelihood
strategies. By collecting and analysing information about whether
and when households changed their decisions regarding land
transfers and off-farm employment, the current and former liveli-
hood strategy type of each household can be determined. By ana-
lysing the type conversion results of the rural households, the
coherent process of rural livelihood transition can be explored.
Finally, according to the SLA, there are several elements related to
people's livelihood strategies. To investigate the effects of house-
hold perceptions and attitudes regarding their specific livelihood
activities and choices, we selected specific factors from these
elements.
3.2. Methodology framework
Our research aims to explore the changes in livelihood strategies
in terms of two types of livelihood activities. The first type is the off-
farm employment activities, which are activities that earn off-farm
income, such as being employed as a migrant worker, working for
township enterprises, operating private industry or working in
commerce. The second type involves land transfer activities, which
are activities related to land transfer-in and transfer-out decisions.
The reason for selecting these livelihood activities is that labour and
land have been two of the most active sources of livelihood capital
in rural livelihood transitions in China over the past 30 years
(Deininger et al., 2014; Fan, 2003; Long et al., 2010; Mullan et al.,
2011).
However, exploring the changes of livelihood strategies is only
one of the objectives of this study. We also explore the factors that
are important to the livelihood transitions. Based on the sustainable
livelihoods framework, a household's assets must be the main el-
ements determining livelihood strategies (Scoones, 1998; Ellis,
2000; Carney, 2003). Access to these assets is mediated by the
policies and institutions, i.e., factors that influence the range of
livelihood options, such as cooperatives, markets, cultures, laws,
government, etc. (Ellis, 2000). Access is also influenced by the
vulnerability context, which are unpredictable factors beyond the
control of households that influence access to an asset (Ellis, 1998).
Such factors include shocks, such as natural calamities and disease
epidemics, trends, such as population trends, economic changes,
technological developments, etc., and seasonality (FAO, 2005).
Together, the livelihood asset base, the policies and institutions,
and the vulnerability context form the basis from which people
develop or pursue their livelihood strategies (Diniz et al., 2013)
(Fig. 2). To explain the changes of rural household livelihood stra-
tegies, it is important to factor in these elements, as is done in this
study.
The procedures used in this research can be subdivided into four
phases (Fig. 3): (1) identification of livelihood strategies: definition
of active livelihood assets and strategies that have changed rapidly
in the past decades; (2) classification of livelihood strategies:
classification of households with different livelihood strategies
based on the identified livelihood assets and strategies; (3)
decision-making factors: discussion of the statistical relationship
between farmers' attitudes and their actual livelihood strategies;
and (4) analyses of livelihood transitions: exploration of the live-
lihood trajectories and the emergence of long-term shifts in live-
lihood strategies.
3.3. Data collection and analysis
3.3.1. Survey approach
To explore changes in rural livelihood strategies and to under-
stand the underlying attitudes and perceptions of households un-
dergoing livelihood transition decisions, a household survey was
used. The survey comprised two groups of questions. The first
group was related to household decisions about off-farm employ-
ment and land transfers. These were used to explore the temporal
characteristics of livelihood transitions in the study region over the
past twenty years. We collected data that referenced the period
from 1993 to 2013 by asking questions such as, “When did you
decide whether to engage in off-farm employment?” and “When
did you decide whether to rent in/out farmland?” The second group
was related to household attitudes about specific livelihood deci-
sion making, such as engaging in off-farm employment and farm-
land transfers. The households were asked to report their attitudes
and perceptions about a number of factors and to evaluate the
importance of these factors in their decisions regarding off-farm
employment and land transfers. A broad range of factors was pro-
posed, which included factors regarding capital and mediating
processes (Table 1). We used the term “attitude score” for the
importance assigned to these factors and coded them on a five-
point scale that ranged from 1 ¼ very weak influence to 5 ¼ very
strong influence. To guarantee accuracy, we did not ask the
 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158 149

Fig. 2. Livelihood strategies and determining factors.

Source: Based on (Ellis, 2000; Diniz et al., 2013; Scoones, 2009)

interviewees about their past opinions due to the potential for

Fig. 3. Methodological framework for livelihood transition analyses.
inaccurate recall or changed perceptions. Moreover, they may be
too young to have participated in past decision making.
A household survey that randomly covered seven villages in the
Qingpu district was conducted in April 2014. First, we interviewed
the village leaders to better understand the existing rural house-
hold conditions, to determine the household typology and to un-
derstand the present proportion of livelihood strategy types (see
Section 4.1). We then randomly chose households (60 households
in each village) based on the previously identified percentages,
using a combined method of face-to-face and telephone surveys.
Finally, we obtained a total of 412 valid samples (eight invalid
samples were removed). The participants selected were generally
the heads of the families. This was ascertained by asking, “Who is
the main decision maker in your family?” the heads of households
were the most representative respondents for this study as they
were the ones who made the household decisions. In China, men
play a more dominant role in families than do women, especially in
rural areas. Although the overwhelming majority of the in-
terviewees were men, a small number of women were also inter-
viewed as they were the heads of their families. The survey content
referred to the basic rural household status and to the time nodes
that indicated when they changed the configuration of their

Table 1
List of factors included in the survey of household attitudes toward livelihood decision making.

Capital/Process Factor Variable

Human (1) Education of head of household Education

(2) Age of head of household Age
(3) Labourers in household Labour
Natural (1) Initially distributed farmland area from the village cooperative DistriAcreage
(2) Actual farmland area currently managed by household CurrentAcreage
(3) Land quality of farmland LandQuality
Physical (1) Agricultural machinery Machinery
Financial (1) Gross income Income
(2) Land rent LandRent
Social (1) Relatives and friends Relatives
Mediating processes (1) Technological assistance Technology
(2) Condition of local infrastructure Infrastructure
(3) Activities organized by local agricultural cooperatives AgriCoop
(4) Policies and subsidies Policy
(5) Crop prices CropPrice
(6) Prices and costs for agricultural material inputs InputPrice
(7) Opportunities in and wages from off-farm employment Wage
150 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158
production, i.e., labour and land.
3.3.2. Data analysis
As the factors that households consider important may not
actually influence their decision making, we used binary logistic
regression models to determine the statistical relationship between
household attitude scores and actual decisions. We extracted the
factors with a statistically significant association to actual decision
making that were driving rural livelihood transitions.
Binary logistic regression models have been widely used for
situations where the dependent variable has a binary output as it is
an ideal model to analyse individual intentions, decision behav-
iours, and influencing factors. In our study, the attitude scores were
used as independent variables, and the actual decisions were used
as dependent variables, allowing for only two possibilities. When
Y ¼ 1, the household has engaged in off-farm employment or has
chosen to transfer land, and when Y ¼ 0, the opposite is true. Given
several independent variables, the probability that Y ¼ 1, written as
P(Y ¼ 1), is given by the following equation:
eðaþb1 X1 þb2 X2 þ/þbk Xk Þ
PðY ¼ 1Þ ¼
1 þ eðaþb1 X1 þb2 X2 þ/þbk Xk Þ
where P(Y ¼ 1) is the probability that Y ¼ 1, e is the base of natural
logarithms, a is the intercept, and bn is the regression coefficient of
the corresponding variable Xn.
The logistic transformation of P(Y ¼ 1) is represented as:
 
PðY ¼ 1Þ
logit ¼ a þ b1 X1 þ b2 X2 þ / þ bk Xk
1  PðY ¼ 1Þ
In the equation, (P(Y ¼ 1)/[(1dP(Y ¼ 1)] denotes the odds that
Y ¼ 1 (the ratio of the probability that Y ¼ 1 to the probability that
Y s 1). Odds ratios [exp (bk)], the natural logarithm of the
regression coefficient, represents the change in the odds of expe-
riencing an event that is affected by an increase of one unit in the
corresponding variable. The significance of each variable was
tested, and the goodness of fit of the models was evaluated using
the Nagelkerke's R2 and the relative operating characteristics (ROC).
The detailed binary logistic method and ROC measurement are
described in more detail by Menard (1995) and Swets (1988).
3.3.3. Classification of livelihood strategies
To analyse rural livelihood strategies, we divided surveyed
households into six groups based on their engagement in each of
the three dominant livelihood activities of the region: off-farm
employment, transferring out their own contracted farmland to
Fig. 4. Household type definition.
others, and transferring in farmland from others. A classification
tree was used to construct the typology (see Fig. 4). Off-farm
employment was defined by whether households engaged in off-
farm employment. Land transfer-out was defined by whether
households rented all of their contracted land to others, and land
transfer-in was defined by whether households rented farmland
from others. By investigating and backtracking the current and
former decisions that households had made to engage in off-farm
employment and land transfer, we obtained the time nodes for
each household specific to when they made their relevant de-
cisions. In this way, the footprint of changes relevant to each
household's livelihood strategies can be restored. The types of
households are dynamic with respect to their livelihood strategy
changes. For example, if a balanced household transferred all its
contracted and transferred land to others and still engaged in off-
farm employment in 2003, it transformed into an off-farm house-
hold. The detailed descriptions of households of each livelihood
strategy type is provided in Section 4.1, and the differentiation
pathways for each household type are analysed in Section 4.3.
4. Results
4.1. Livelihood typology
4.1.1. Off-farm employment-oriented household (OFO)
Households of this type distribute part of their labour into off-
farm employment and maintained their originally contracted
farmland for agricultural production. They do not rent farmland
from others, and the main source of their income is derived from
off-farm employment. For households of this type, young male
labourers are generally involved in off-farm employment, whereas
women and older men engage in on-farm activities. Under the
background of fast urbanization in Shanghai, households of this
type are typically in a state of rapid differentiation and tremendous
instability. When better conditions, such as better land transfer
markets, social security, education for children and health care for
migrant workers, are met, differentiation is more likely to occur.
4.1.2. Balanced household (BA)
Households of this type combine on-farm and off-farm
employment as their sources of income. They engage in off-farm
employment, while at the same time, renting farmland from
others. Generally, their labour forces are relatively plentiful enough
to engage in off-farm employment and comparatively small scale
farm management simultaneously. The relatively balanced income
sources of this household type suggest that these households
 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158
demonstrate greater productive vitality and aspirations of higher
income. Meanwhile, they also exhibit a higher capacity for resource
allocation and capital accumulation.
4.1.3. Off-farm household (OF)
Households of this type allocate all of their labour forces to off-
farm employment and rent out all of their contracted farmland to
others, thus, they no longer engaged in farming activities. There are
two types of off-farm employment, that which involves unskilled
labour and that which requires specific skills and knowledge. The
relatively high job stability and income levels make the latter more
capable of transforming workers into urban residents, while those
involved in unskilled labour are more likely to return to agricultural
production as they get older. Many households of this type are not
dependent on farmland, and they have been independent of agri-
cultural economics with the exception that their homesteads and
registrations are still in the village.
4.1.4. Idle household (IDL)
Households of this type have rented out all of their contracted
farmland to others, while engaging in neither farm nor off-farm
employment due to, for example, old age or disease. In general,
these households are characterized by old-age, lack of work and
poverty. The main source of their income is land rent, subsidies
from the government and/or old-age insurance.
4.1.5. Professional household (PRO)
Households of this type do not engage in off-farm employment.
They rent farmland from others to realize the appropriate scale of
management and engage all of their labour forces in farming ac-
tivities. They are characterized by large scale farming and sole
agricultural income sources.
Professional households are often called specialized households
or family farms in China. They are characterized by better mana-
gerial capacity and investment foresight. They have divested from
the traditional mode of agricultural production and transformed
their farms into market-oriented operations with high market ra-
tionality of production investment, technology adoption, and
product sales. Because they require a larger number of one-off
deposits and investments when they transfer in a substantial
amount of farmland, there is a greater demand from the house-
holds for capital accumulation. As the main farming operators,
professional households, which are currently supported by the
Shanghai government, have had a positive effect on agricultural
development in China and will be the dominant force of modern
agriculture. It is worth noting that a large portion of heads of
households of this type have been or are village leaders. In general,
they have more sophisticated agricultural techniques and are more
knowledgeable about the government's policies with respect to
agricultural support.
4.1.6. Conventional household (CON)
Households of this type do not engage in off-farm employment,
and they do not rent farmland from others. However, they do retain
all the features of conventional rural-households. In general,
restricted by the resource endowment of farmland, households of
this type have a lower income. Lacking access to gain off-farm in-
come, they are highly dependent on farmland self-sufficiency,
which typically results in poverty. At present, households of this
type have little influence over agricultural production efficiency
and output in the region.
4.2. Decision-making factors
In Table 2, the factors considered important for livelihood
151
decision making in the study area are ranked in order of their
average attitude scores. The logistic results that relate the actual
decisions to the attitude scores are also presented in Table 2. The
models of off-farm employment and land transfer-in and transfer-
out exhibit a good fit as the values of Nagelkerke's R2 and the ROC of
these models are relatively good, thus indicating that the attitude
scores represent significant relationships with these decisions.
Factors with significant coefficients (marked in bold in Table 2)
are primarily located at the top of the ranking list, which indicates
that most of the factors that households perceive as important also
play a role in actual livelihood decision making. The logistic results
indicate that two capital factors, Age and Education and two
mediating factors, Wage and CropPrice, significantly influence off-
farm employment decision making. The odds ratio of wage and
education are above 1, indicating that households that consider off-
farm employment opportunities, income, and education of the
heads of households as important decision factors are more likely
to engage in off-farm employment. The odds ratio for Age and
CropPrice (between 0 and 1) suggests the opposite effect.
The land transfer-in decision exhibits a positive correlation with
attitude scores with respect to Policy, Machinery, and AgriCoop
factors. The land transfer-out decision exhibits a positive correla-
tion with attitude scores for LandRent and Wage factors. At the
same time, this decision demonstrates a negative correlation with
the attitude scores for Labour, indicating that households that
consider labour important to their decisions are less inclined to
make a positive decision on land transfer-out.
4.3. Livelihood transitions
Based on the variations in livelihood strategy trends from 1993
to 2013, we analysed and explored the transition processes of each
household type according to three specific periods. The results are
presented in Fig. 5, Tables 3e5. Fig. 5 presents the changing trends
in numbers of households with respect to various livelihood
strategy types during each of the three periods, namely, 1993 to
1999, 1999 to 2005, and 2005 to 2013. Tables 3e5 displays the
transferring quantity between various household types for the
three periods. For each period, the number at the end of each col-
umn indicates the number of households of each type at the
beginning year, and the numbers at the end of each line represent
the number of households of each type at the ending year.
4.3.1. Off-farm employment-oriented households
Between 1993 and 1999, the number of households of this type,
which were often converted from conventional households, rapidly
increased from 198 to 235 (Fig. 5A, Table 3). Under the distinct land
tenure system, households risked losing their rights to farmland if
they permanently migrated, a loss that increased the risk of losing
basic living guarantees and future income (Mullan et al., 2011).
Meanwhile, the transaction cost of market-based land transfers was
high due to the fragmentation of the land and the insecurity of
property rights (Deininger et al., 2014). In these situations, the
instability of off-farm employment impeded land transfer-out ac-
tivities for most of the households, having transformed many
conventional households into off-farm employment-oriented ones.
At the beginning of the 1999 to 2005 period, the number of off-
farm employment-oriented households remained stable as the off-
farm employment of surplus labour forces reached equilibrium
(Fig. 5A). During the latter part of this phase, however, a portion of
the off-farm employment-oriented households became off-farm
households, and this number began to stabilize (Fig. 5A, Table 3).
During this period, as the extensive use of farm machinery signif-
icantly improved the agriculture productivity in Shanghai, certain
requirements for farmland transfers were established. On the other
152 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158

Table 2
The average attitude scores on off-farm employment and land transfer, and the estimated coefficients of the logistic model relating household attitude scores and actual
decisions [exp (b) values in logistic model].

Off-farm employment Land transfer

Variable Average score Exp (b) values in logistic model Variable Average score Exp (b) values in logistic
model

Transfer-in Transfer-out

1 Wage 3.62 1.581** LandRent 3.07 2.36 0.696**

2 Income 3.56 1.218 Policy 2.91 1.576** 0.495
3 Age 3.27 0.685** Wage 2.76 1.215 0.438*
4 LandRent 3.14 1.425 Machinery 2.69 1.215** 1.275
5 Labour 3.07 1.198 CropPrice 2.56 0.671 0.183
6 Education 2.55 1.358* Income 2.48 1.269 1.021
7 CurrentAcreage 2.51 1.339 AgriCoop 2.39 0.858** 0.755
8 CropPrice 2.43 0.627* Technology 2.27 1.81 1.912
9 Relatives 2.41 0.762 CurrentAcreage 2.15 1.233 0.487
10 Infrastructure 2.35 1.025 Labour 2.09 0.95 0.496*
11 Policy 2.27 0.947 InputPrice 1.93 3.066 3.525
12 LandQuality 2.17 0.753 Infrastructure 1.86 0.145 0.835
13 InputPrice 1.97 0.934 LandQuality 1.82 1.74 0.902
14 Machinery 1.83 0.794 Education 1.72 0.661 1.308
15 Technology 1.73 1.014 Relatives 1.59 0.45 1.231
16 DistriAcreage 1.51 0.868 DistriAcreage 1.38 1.057 1.683
17 AgriCoop 1.01 0.759 Age 1.35 0.167 2.203
ROC statistic 0.857 ROC statistic 0.892 0.899
Nagelkerke R square 0.361 Nagelkerke R square 0.54 0.523

* Significance at p < 0.1 level.

** Significance at p < 0.05 level.
Factors with significant coefficients in the logit results are marked as bold.

hand, the 1998 revision of the Land Administration Law (LAL) and
the 2003 Rural Land Contract Law (RLCL) improved the tenure se-
curity and transferability of rural lands (Deininger et al., 2014).
Moreover, the 1998 revision of the LAL stipulated that (i) the village
collective should compose written land contracts with individual
households and (ii) the duration of land use rights would extended
by another 30 years (Ma et al., 2015). The RLCL focused on the
following three areas: (i) defining land rights as property rights
rather than just private contracts, (ii) prohibiting land reallocation
within villages in response to demographic changes and limiting
partial land readjustments with clear conditions, (iii) specifying the
exact conditions and modes under which farmers can transfer
arable land (Ma et al., 2015). Under these conditions, households
whose off-farm employments were relatively stable and whose off-
farm income was relatively high rented out their contracted land.
However, most of the households were still unwilling to rent out
their contracted land. One reason for this was because the
remaining labourers who were not absorbed by off-farm industries,
were distributed among the households, and thereby tightly bound
to the contracted land. In addition, the relatively lower land rent
during this phase did not inspire households to rent out their land.
Moreover, the dependency on farmland caused some of the
households to hesitate to rent out their land even if they had stable
off-farm income and had the opportunity to move their families
into town.
Between 2005 and 2013, off-farm employment-oriented
households quickly diminished as they transformed into off-farm
households (Fig. 5A, Table 3).
4.3.2. Balanced households
According to our surveys, between 1993 and 1999, few balanced
households existed (Fig. 5B).
However, between 1999 and 2005, the number of balanced
households began to slowly increase (Fig. 5B, Table 4). During this
period, the abolition of the agricultural tax and the increase in grain
planting subsidies slightly improved household incomes. During
2000 and 2003, the central government implemented the tax-for-
fee reform (TFR), which abolished local fees levied on rural
households (Wang and Shen, 2014). In 2002 and 2003, the agri-
cultural tax in Shanghai was gradually abolished, thus reducing the
annual burden on peasants by 4.25 billion Yuan. In 2004, China
embarked on the grain subsidy programme, which included direct
food subsidies, direct agricultural material subsidies, selective seed
subsidies, and farm machinery purchase subsidies (Yi et al., 2015).
The amount of grain subsidies given to farmers increased from 14.5
billion Yuan in 2004 to 166.8 billion Yuan (Chen, 2013) in 2012.
Implementing the grain subsidy programme improved farm in-
come. Meanwhile, widespread agricultural machinery significantly
decreased labour force investment intensity with respect to
ploughing, sowing, and reaping, thereby inducing some households
to expand their agricultural production scale. Surveys indicate that
land conversions during this period were usually informal, were
often verbal, and were frequently executed with relatives.
Between 2005 and 2013, the number of households of this type
first increased and then declined (Fig. 5B). During this period,
balanced households were mainly transformed from conventional
households and off-farm employment-oriented households
(Table 4). The results of the survey further indicate that at the
beginning of the phase, with the encouragement of a moderate
scale management policy, the number of balanced households
significantly increased. The moderate scale management policy has
been pursued by the Shanghai government since 2006, and the
related policy provisions focused on two areas, namely, (i)
normalizing and regulating the form and the content of land
transfers such that transfer contracts adhere to unified standards
and be intermediated by collectives and minimum standards of
land rent be updated by the government on an annual basis; and (ii)
providing cash subsidies to moderate scale tenant farmers ac-
cording to the acreage of land they rent. During the middle and final
periods of this phase, because households of this type combined
on- and off-farm employment as their sources of income, the
dispersion of the labour force caused the majority of these house-
holds to operate on a relatively smaller scale. However, smaller
scale production could not absorb the constantly rising rents, and
 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158 153

Fig. 5. Differentiation trends of various types of households between 1993 and 2013.

Table 3 Table 5
Transition matrix of the different household types between 1993 and 1999. Transition matrix of the different household types between 2005 and 2013.

1999 2013

OFO BA OF IDL PRO CON Total OFO BA OF IDL PRO CON Total

1993 OFO 179 0 0 0 1 55 235 2005 OFO 12 0 0 0 0 14 26

BA 2 2 0 0 0 2 6 BA 7 5 1 0 0 0 13
OF 15 0 24 0 0 0 39 OF 164 8 106 0 2 27 307
IDL 0 0 0 0 0 2 2 IDL 0 0 0 3 0 18 21
PRO 2 0 1 0 0 1 4 PRO 8 9 7 0 5 0 29
CON 0 0 0 0 0 126 126 CON 0 0 0 0 0 16 16
Total 198 2 25 0 1 186 412 Total 191 22 114 3 7 75 412

Table 4 the households could not actualize the benefits of mechanized

Transition matrix of the different household types between 1999 and 2005. farming. Meanwhile, the increase in non-agricultural job oppor-
2005 tunities and incomes prompted households to make trade-offs. In
OFO BA OF IDL PRO CON Total
this case, the balanced households operating at a relatively smaller
scale gradually disappeared.
1999 OFO 154 0 0 0 0 37 191
BA 9 5 2 0 0 6 22
OF 70 0 35 0 1 8 114 4.3.3. Off-farm households
IDL 0 0 0 2 0 1 3 Between 1993 and 1999 there were few off-farm households
PRO 2 0 2 0 3 0 7
and little variation (Fig. 5C).
CON 0 1 0 0 0 74 75
Total 235 6 39 2 4 126 412 However, between 1999 and 2005, the number of off-farm
households began to slowly increase before stabilizing (Fig. 5C).
154 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158
Surveys indicate that these off-farm households were mainly
transformed from off-farm employment-oriented households
(Table 4). After 1999, farmlands retained as off-farm employment-
oriented households gradually declined due to the widespread use
of agricultural machinery, the steady increase of agricultural pro-
ductivity and farmland rent, thus generating a number of off-farm
households.
Between 2005 and 2013, the number of households of this type
increased rapidly and then stabilized (Fig. 5C). These households
had been transformed from off-farm employment-oriented
households (Table 5). The economies of scale and normalization
brought by the regular circulation of farmlands gradually became
institutional and technological guarantee for land transfer. Mean-
while, the land rental income was even higher than the farming
income itself, causing most of the off-farm employment-oriented
households to rent out their farmlands, thus transforming them
from on-farm households to off-farm households.
4.3.4. Idle households
Between 1993 and 1999, households of this type were few
(Fig. 5D), and between 1999 and 2005, the number of idle house-
holds only minimally increased and exhibited a fluctuating trend
(Fig. 5D).
However, between 2005 and 2013, the number of idle house-
hold, which were mainly transformed from conventional house-
holds, increased rapidly (Fig. 5D) (Table 5). Surveys indicate that
under the massive increase in demand and in the rent payments for
farmlands, the households that regarded farming as negative
gradually disengaged from agricultural production.
4.3.5. Conventional households
Between 1993 and 1999, the number of conventional house-
holds quickly declined from 186 to 126 (Fig. 5E). The official
establishment of China's market economic system in 1993 drove a
large number of rural surplus labour forces to emerge as migrant
workers, thereby driving the number of conventional households
into a sustained downward trend. It is worth noting that, in the
cases of extremely low earnings from farming, a number of the
farmlands were abandoned.
Between 1999 and 2005, the number of conventional house-
holds continued to decrease and then began to stabilize (Fig. 5E).
There were two major reasons for this. First, the impact of the Asian
financial crisis of 1997 on employment first impacted the rural
migrant workers with less job stability, causing some of these
workers to return home. Second, around 1998, the land contract of
the first round expired, and the second round began. The fear of
losing farmland because of migrant employment prompted some
rural migrant workers to return home.
Between 2005 and 2013, the number of households of this type
continued to decline (Fig. 5E). From 2006, under the strong support
of the Shanghai municipal government, the subsidies for moderate
scaled farming increased year by year, and the land transfer rents
increased substantially. This increase allowed the transfer rents to
cover, for the most part, the household farming income, which
greatly increased the household motivation to transfer their
farmlands from production.
4.3.6. Professional households
Between 1993 and 1999, few professional households were
identified (Fig. 5F). During this phase, under the rapid growth of
production costs, household enthusiasm for agricultural produc-
tion was low. The low income generated from grain crop cultivation
resulted in a lack of household profit. This lack of profit served as a
motive for scaled management, which restrained the demand for
household farmland and the development of a land transfer
market. Meanwhile, fragmentation and decentralization, caused by
the principle of absolute equal distribution, limited contiguous
planting on a large scale when households themselves transferred
farmland.
Between 1999 and 2005, there was still a minimal number of
professional households (Fig. 5F), and the surveys indicated that
households of this type had relatively small production scales
during this period.
Between 2005 and 2013, the number of professional households
quickly increased and began to stabilize (Fig. 5F). The surveys
indicated that households of this type were mainly transformed
from balanced and off-farm households and had relatively large
production scales during this period (Table 5). Following a short
period of off-farm engagement, the households accumulated a
certain amount of capital to sustain a larger number of one-off
deposits and investments when they transferred a substantial
amount of farmland. After the central government cancelled agri-
cultural taxes and began to implement a national policy of direct
subsidies for grains, the enthusiasm of farmers to grow grain
mobilized. Meanwhile, in 2006, Shanghai began to establish sub-
sidies for large-scale agricultural operations and the purchase of
farm machinery. The increasing subsidies given to households
whose cultivated farmlands reached a certain area created a
distinct increase in the number of professional households. The
surveys indicate that the attitudes of the village collectives
regarding land transfers and the development of intermediaries
were important variables in farmland transfers.
5. Discussion
5.1. Classification of livelihood strategies
According to the definition of livelihood strategy, several studies
have classified livelihood strategies based on a number of criteria
(see, e.g., Daskalopoulou and Petrou, 2002; Jansen et al., 2006;
Pichon, 1997; Shackleton et al., 2001; Soltani et al., 2012; Tittonell
et al., 2010). However, some of the criteria are understood as out-
comes of livelihood strategies, e.g., income composition and land
area, while others are difficult to comprehend in the context of
exploring the processes of livelihood transitions, e.g., land use share
and crop selection. In this study, the combination of decisions, as
made by rural households, regarding off-farm employment and
land transfers, serve as the criteria for classification. One reason for
this is that these decisions are the most important variables influ-
encing the livelihood of rural households in China over the past 20
years. Moreover, as these decisions are easily recalled with greater
accuracy by interviewees, we could more successfully explore the
processes of their livelihood transitions.
5.2. Driving forces
This paper used attitude score to identify the key factors influ-
eincing the decisions of rural households. In the binary logistic
models, factors from livelihood capital and mediating processes are
proposed as independent variables. The regression results indicate
that several factors identified by households as important to deci-
sion making (household attitude) have a significant relationship
with the actual decisions of the households. These factors provided
some insight into the mechanism of livelihood transitions.
Although this approach successfully extracted the factors that are
important for household decision making, it also contained certain
treatments that make the analyses straightforward. For example, as
the attitude scores used as the independent variables present only
the households' current opinions, they inevitably overlook the
dynamics of household attitudes. Moreover, the method to
 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158
measure household attitudes is qualitative. Therefore, future
research must develop more accurate ways to estimate household
perceptions and attitudes.
5.3. Livelihood transition stages
The continuously changing processes of livelihood strategies
have not been addressed in previous studies (Mushongah and
Scoones, 2012). Thus, we have conducted a more thorough anal-
ysis of livelihood transitions. Combining the characteristics of
livelihood transition of each household type, we conclude that the
trajectory of rural household livelihood transitions has progressed
through three stages from 1993 to 2013, specifically, the initial
stage, the household self-organizing stage, and the government-led
stage.
The period 1993 to 1999 constituted the initial stage of the
livelihood transition. The establishment of a market economic
system and the rapid development of township enterprises
prompted a large number of rural surplus labourers to move into
non-agricultural industries. Although there was a legal basis for
farmland transfers, the demand for farmland was depressed
because of the low comparative gains of agricultural production,
thereby resulting in a low rate land transfers. At this stage, while
many off-farm employment-oriented households appeared, there
were few balanced and professional households. The off-farm
employment of rural labourers and the process of household off-
farm transformation was not synchronous. With the differentia-
tion of conventional households into off-farm employment-ori-
ented households as the main characteristic, this was the initial
stage of the livelihood transition.
The period from 1999 to 2005 was the self-organizing stage of
the livelihood transition. The labour productivity of households
quickly improved under the application of agricultural machinery.
In addition, the land administrative laws, which were modified in
1998, officially prescribed a land contracting system that would not
change for 30 years. The policy of the stabilization of peasant land
contract rights was legally binding, thus promoting an increased
demand for land transfers. During this phase, land transfers with
smaller scales and informal forms were spontaneous between
households, and those households that rented in farmlands were
mainly balanced ones. It is also noted that land rent was slowly
increasing. For example, surveys indicate that the land rent was
approaching zero in 1999 but that it slowly increased over the next
several years, albeit remaining at a low level. This stage was char-
acterized by the appearance of a small number of balanced and
professional households, while some of the off-farm employment-
oriented households transformed into off-farm households that
tended toward stability around 2005.
The period from 2005 to 2013 was the government-led stage of
the livelihood transition that was characterized be the trans-
formation of many off-farm employment-oriented households to
off-farm households and the tapering off of balanced households as
the number of professional households quickly increased and
tended towards stability. After 2006, using the methods of scaled
operation subsidies, agricultural extension services and normali-
zation of land transfers, the Shanghai government vigorously pur-
sued the development of the moderately scaled operation of
agriculture, which resulted in the constant increase in land trans-
fers, the expansion in the scale of production and the increase in
farmland rent. At the end of this stage, the ratio of land transfers
reached 93%, and the household differentiation and livelihood
transitions were tending toward stability. The professional house-
holds gradually displaced the balanced households as the major
agents of moderately scaled operations in agriculture such that
there was the formation of the coexistent pattern of many off-farm
155
and professional households.
5.4. Policy implications
Comparing related studies from different regions of China, it is
noted that the processes of rural livelihood transition have both
similarities and differences (Feng, 2008; Feng and Heerink, 2008;
Gao et al., 2012; Yan et al., 2014). The two main phenomena
emerge during the process of rural livelihood transition in the
Qingpu district, the movement of labour from agriculture and the
development of the land transfer market, all of which, though
universal in China, display various development situations in
different regions (Huang et al., 2012; Jin and Deininger, 2009; Van
den Berg et al., 2007). As China's land transfer market was in an
emerging phase (Tian and Li, 2014), the appearance of land rental
markets was not conspicuous until the late 1990s. In the late 1980s
and early 1990s, few farmers engaged in rental activities (Brandt
et al., 2004; Deininger and Jin, 2005). After the promotion of the
RLCL, however, land rental activities began to expand(Yan et al.,
2014). The problems appeared in the rural development of
Qingpu. For example, the low frequency of farmland transfers,
which was out of sync with the high frequency of household off-
farm employment, and the individual and temporary migration
strategies, which hampered the rural allocative efficiency and the
rate of urbanization in China, that most of the off-farm employ-
ment-oriented households were taking, were also observed and
explored in other studies on China's rural development.
The similarities are due to the relatively coincident context of
China's policy and institutional arrangements and its socio-
economic development environment as well as the differentia-
tion, which can be explained by regional distinctions, such as the
level of economic development, the natural resource conditions,
the administrative abilities of local governments, and the local
traditional culture. For example, even though Chinese authority has
implemented uniform land laws and policies, different regions have
their own land institutions that have their own rules concerning
land distribution, adjustments, and transfers. There is also spatial
unevenness in the way in which land laws and policy reforms have
been implemented (Feng and Heerink, 2008). In some of the more
economically developed provinces, such as Zhejiang, one-third of
all farm land is being rented (Kung, 2002). According to a nation-
wide survey, Gao et al. (2012) found that, by 2008, 19% of the
cultivated land was rented. However, most of the rental contracts
were verbal and informal and frequently involved relatives
(Deininger et al., 2014; Gao et al., 2012).
The rural livelihood transition in the Qingpu district, by contrast,
was in-depth and occurred quite rapidly. The percentage of the
number of off-farm and professional households in 2013, according
to our survey, was relatively high (93%). While the relatively
smooth transition from a rural livelihood is the result of several
factors, the policy aspects are the focus of this study. It is noted that
the rural livelihood transition in Shanghai is closely related to the
highly developed economy of Shanghai, which has provided suffi-
cient opportunities for income, a factor that has drawn rural
labourers from the agricultural sector, while the topography of
Shanghai is well suited for the large-scale farming mechanization
of professional households. However, the central and local gov-
ernments play a crucial role in doing away with inefficient insti-
tutional arrangements and in erecting policy arrangements that can
improve the allocation efficiency of rural household livelihood as-
sets, thereby shaping the trajectory of rural livelihood transition.
The fact that, in this study, the transfers of rural labour to off-farm
jobs, which were active for at least a decade and which did not spur
the development of land markets, attest to this. During this transfer
period, land is trapped in the off-farm employment households
156 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158
because of the instability of off-farm employment, the insecurity of
land tenure, the high transaction costs of rural land transfers and
the low comparative benefits of agricultural production.
Policy and institutional arrangements with respect to the land
and labour markets form a basis of the transfer of labour from the
agricultural sector to higher-paying, non-agricultural pursuits and
from the market-based land transfers to more productive farmers
then increase farm size and total earnings (Deininger et al., 2014).
China's policies addressing these problems can benefit other
developing countries as these countries make policy decisions
regarding the security and transferability of rural land through law
provisions and institutional arrangements, the support provided to
non-migrants as they adjust their agricultural land holdings, the
increase in the comparative gains for agricultural production
through subsidizing agricultural land and the improvements of the
rural infrastructure and living conditions of farmers.
6. Conclusions
Rural households in China are changing their livelihood strate-
gies by redeploying their livelihood assets, of which labour and
land are among the most critical. The labour demand, which was
triggered by the rapid development of non-agricultural sectors,
accompanied by the large surplus of labour in the agricultural
sector and caused by the development of modern technology,
resulted in a large-scale movement of labour from agriculture in the
pursuit of higher income. The land left behind, which was small and
fragmented and characterized by tenure insecurity, offered an
extremely low comparative benefit, and was in a state of inefficient
use. This compelled the government to develop a series of policy
arrangements designed to improve the allocative efficiency of the
land. In response to the new opportunities presented by economic
development and institutional improvement, rural households are
adjusting their livelihood strategies. The process of rural household
livelihood transition constitutes the epitome of local rural devel-
opment during the past decades. Accordingly, to a large extent, the
nature and speed of this process are dependent on the policy and
institutional environment. Examining the same households over
time allows a better understanding of the conditions that keep
people in poverty as well as a more comprehensive perceptiona of
what enables these individuals to inform policy and improve their
situation (Kristjanson et al., 2010).
Using the sustainable livelihood framework, this paper adds to
the current literature by conducting a structured analysis of rural
household livelihood transitions in the context of the Chinese land
tenure system. This case study explores two important character-
istics of rural household livelihood transitions, namely, off-farm
employment and land transfers, and the drivers of these transi-
tions in east costal China by using a household survey-based
approach. Such a longitudinal study has several advantages that
could be extended to future studies of rural livelihood dynamics.
First, the analysis goes beyond the traditional relations between
regional external or household asset factors and livelihood strategy
changes by analysing the ways in which perceptions of farmers
influence their actual decision making. The survey results indicate
that two capital factors, age and education, and two mediating
factors, wage and crop prices, are important for making off-farm
employment decisions, whereas factors such as policies and sub-
sidies, machinery, activities organized by local agricultural co-
operatives and land rent substantially influence land transfer
decisions.
Second, based on household decision making regarding off-farm
employment and land transfers, the households were divided into
six types, namely, off-farm employment-oriented, balanced, off-
farm, idle, professional, and conventional. Our surveys of the
Shanghai Qingpu district allowed us to analyse the evolutionary
characteristics, developing trends and driving factors of the pro-
cesses of rural household livelihood transitions between 1993 and
2013. The results indicated that the overall trend of the differenti-
ation with respect to household livelihood was such that the pro-
portion of conventional households decreased and the proportion
of off-farm employment-oriented households and off-farm
households alternately increased. The main proprietors of agricul-
tural production transformed from scattered households, i.e., con-
ventional and off-farm employment-oriented households, to large-
scale management households, i.e., professional and balanced
households. This trend constituted the primary characteristics of
rural household livelihood transitions during the process of fast
industrialization combined with rapid urbanization.
Finally, rural livelihood transitions in Qingpu followed a stepped
periodic characteristic that included an initial stage, a self-
organizing stage and a government-led stage. In the process of
livelihood transition, the peasant self-organizing and government-
led mechanisms should act synergistically. Thus, the livelihood
transitions under peasant self-organization would sometimes
encounter stagnation characterized by inefficiency and lack of vi-
tality. Therefore, to break through the obstacles in time, the gov-
ernment should be sensitive to the inefficient case and improve the
policies and the institution. In the case of rural livelihood transition
in rural Qingpu, the key policy arrangements are grouped into two
categories. One category is the policies to increase the security and
transferability of rural land, and the other is the policies to raise the
comparative gains for agricultural production. As Shanghai is at the
forefront of China's market reforms and economic developments,
the method and results of this study are expected to have some
significance for similar studies, such as those in northern and
central China where similar processes of livelihood transitions are
still in the early stages.
Acknowledgements
This research was supported by the National Natural Science
Foundation of China (No. 41130526). The authors thank all of the
farmers who gave their time to be interviewed for this study. I offer
many thanks to the anonymous reviewers who provided excellent
feedback to improve the quality of this manuscript.
References
Addison, T., Hulme, D., Kanbur, R., 2009. Poverty dynamics: measurement and
understanding from an interdisciplinary perspective. In: Addison, T., Hulme, D.,
Kanbur, R. (Eds.), Poverty Dynamics: Interdisciplinary Perspectives, p. 3e26.
Bhandari, P.B., 2013. Rural livelihood change? Household capital, community re-
sources and livelihood transition. J. Rural. Stud. 32, 126e136.
Brandt, L., Rozelle, S., Turner, M.A., 2004. Local government behavior and property
right formation in rural China. J. Instit. Theor. Econom.-Z. Fur Die Gesamte
Staatswiss. 160, 627e662.
Carney, D., 2003. Sustainable Livelihoods Approaches: Progress and Possibilities for
Change. Department for International Development (DFID), London, UK.
Chambers, R., Conway, G.R., 1992. Sustainable Rural Livelihoods: Practical Concepts
for the 21st Century. Discussion Paper. Institute of Development Studies, Uni-
versity of Sussex, p. 42.
Chang, H., Dong, X.-Y., Macphail, F., 2011. Labor migration and time use patterns of
the left-behind children and elderly in rural China. World Dev. 39, 2199e2210.
Chen, F., Davis, J., 1998. Land Reform in Rural China since the Mid-1980s. FAO, Rome,
Italy. ftp://ftp.fao.org/SD/SDA/SDAA/Lr982/art-9.pdf.
Chen, H., Zhu, T., Krott, M., Calvo, J.F., Ganesh, S.P., Makoto, I., 2013a. Measurement
and evaluation of livelihood assets in sustainable forest commons governance.
Land Use Policy 30, 908e914.
Chen, H., Zhu, T., Krott, M., Maddox, D., 2013b. Community forestry management
and livelihood development in northwest China: integration of governance,
project design, and community participation. Reg. Environ. Chang. 13, 67e75.
Daskalopoulou, I., Petrou, A., 2002. Utilising a farm typology to identify potential
adopters of alternative farming activities in Greek agriculture. J. Rural. Stud. 18,
95e103.
Deininger, K., Jin, S.Q., 2005. The potential of land rental markets in the process of
 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158
economic development: evidence from China. J. Dev. Econ. 78, 241e270.
Deininger, K., Jin, S.Q., Xia, F., 2014. Moving off the farm: land institutions to facil-
itate structural transformation and agricultural productivity growth in China.
World Dev. 59, 505e520.
DFID, 1999. Sustainable Livelihood Guidance Sheets. DFID, London, UK.
Diniz, F.H., Hoogstra-Klein, M.A., Kok, K., Arts, B., 2013. Livelihood strategies in
settlement projects in the Brazilian Amazon: determining drivers and factors
within the Agrarian Reform Program. J. Rural. Stud. 32, 196e207.
Dong, X.Y., 1996. Two-tier land tenure system and sustained economic growth in
post-1978 rural China. World Dev. 24, 915e928.
Ellis, F., 1998. Household strategies and rural livelihood diversification. J. Dev. Stud.
35, 1e38.
Ellis, F., 2000. Rural Livelihoods and Diversity in Developing Countries. University
Press Oxford, UK.
Fan, C.C., 2003. Rural-urban migration and gender division of labor in transitional
China. Int. J. Urban Reg. Res. 27, 24.
Fang, Y.-p., 2013. The effects of natural capital protection on pastoralist's livelihood
and management implication in the source region of the Yellow River, China.
J. Mt. Sci. 10, 885e897.
Fang, Y.-p., Fan, J., Shen, M.-y., Song, M.-q., 2014. Sensitivity of livelihood strategy to
livelihood capital in mountain areas: empirical analysis based on different
settlements in the upper reaches of the Minjiang River, China. Ecol. Indic. 38,
225e235.
FAO, 2005. Rapid Guide for Missions: Analyzing Local Institutions and Livelihoods.
Viale delle Terme di Caracalla, Rome, Italy. ftp://ftp.fao.org/docrep/fao/008/
a0273e/a0273e00.pdf.
Feng, S., 2008. Land rental, off-farm employment and technical efficiency of farm
households in Jiangxi Province, China. Njas-Wageningen J. Life Sci. 55, 363e378.
Feng, S., Heerink, N., 2008. Are farm households' land renting and migration de-
cisions inter-related in rural China? Njas-Wageningen J. Life Sci. 55, 345e362.
Galipeau, B.A., Ingman, M., Tilt, B., 2013. Dam-induced displacement and agricul-
tural livelihoods in China's Mekong Basin. Hum. Ecol. 41, 437e446.
Gao, L.L., Huang, J.K., Rozelle, S., 2012. Rental markets for cultivated land and
agricultural investments in China. Agric. Econ. 43, 391e403.
Groom, B., Palmer, C., 2012. REDDþ and rural livelihoods. Biol. Conserv. 154, 42e52.
Guo, X., Zhou, X., 2013. Livelihood and development of older generation of rural
migrant workers after returning home. Chin. Rural. Econ. 53e62.
Herrmann, S., Fox, J.M., 2014. Assessment of rural livelihoods in South-West China
based on environmental, economic, and social indicators. Ecol. Indic. 36,
746e748.
Hogarth, N.J., Belcher, B., 2013. The contribution of bamboo to household income
and rural livelihoods in a poor and mountainous county in Guangxi, China. Int.
For. Rev. 15, 71e81.
Hogarth, N.J., Belcher, B., Campbell, B., Stacey, N., 2013. The role of Forest-Related
income in household economies and rural livelihoods in the Border-region of
southern China. World Dev. 43, 111e123.
Hu, W., 1997. Household land tenure reform in China: Its impact on farming land
use and agro-environment. Land Use Policy 14, 175e186.
Huang, J.K., Gao, L.L., Rozelle, S., 2012. The effect of off-farm employment on the
decisions of households to rent out and rent in cultivated land in China. China
Agric. Econ. Rev. 4, 5e17.
Huber, F.K., Yang, Y., Weckerle, C.S., Seeland, K., 2014. Diversification of livelihoods
in a society in transition: a case study of Tibetan Communities in Southwest
China. Soc. Nat. Resour. 27, 706e723.
Jansen, H.G.P., Pender, J., Damon, A., Wielemaker, W., Schipper, R., 2006. Policies for
sustainable development in the hillside areas of Honduras: a quantitative
livelihoods approach. Agric. Econ. 34, 141e153.
Jin, S.Q., Deininger, K., 2009. Land rental markets in the process of rural structural
transformation: productivity and equity impacts from China. J. Comp. Econ. 37,
629e646.
Jongeneel, R.A., Polman, N.B.P., Slangen, L.H.G., 2008. Why are Dutch farmers going
multifunctional? Land Use Policy 25, 81e94.
Kemp, D.R., Han, G., Hou, X., Michalk, D.L., Hou, F., Wu, J., Zhang, Y., 2013. Innovative
grassland management systems for environmental and livelihood benefits.
Proc. Natl. Acad. Sci. U. S. A. 110, 8369e8374.
Kristjanson, P., Mango, N., Krishna, A., Radeny, M., Johnson, N., 2010. Understanding
poverty dynamics in Kenya. J. Int. Dev. 22, 978e996.
Kung, J.K.S., 2002. Off-farm labor markets and the emergence of land rental markets
in rural China. J. Comp. Econ. 30, 395e414.
Li, C., Liu, W., Feng, W., Li, S., 2013. Influence of relocation policy on farmer
households' livelihood strategy. China Rural. Surv. 31e44.
Li, G., Fang, C., Qiu, D., Wang, L., 2014. Impact of farmer households' livelihood assets
on their options of economic compensation patterns for cultivated land pro-
tection. J. Geogr. Sci. 24, 331e348.
Liao, C., Sullivan, P.J., Barrett, C.B., Kassam, K.-A.S., 2014. Socioenvironmental threats
to pastoral livelihoods: risk perceptions in the Altay and Tianshan Mountains of
Xinjiang, China. Risk Anal. 34, 640e655.
Liu, J., 2014. Ageing, migration and familial support in rural China. Geoforum 51,
305e312.
Liu, Y., Chen, Y., Long, H., 2011. Regional diversity of peasant household response to
new countryside construction based on field survey in eastern coastal China.
J. Geogr. Sci. 21, 869e881.
Liu, Y., Yang, R., Long, H., Gao, J., Wang, J., 2014. Implications of land-use change in
rural China: a case study of Yucheng, Shandong province. Land Use Policy 40,
111e118.
157
Long, H., Li, Y., Liu, Y., Woods, M., Zou, J., 2012. Accelerated restructuring in rural
China fueled by 'increasing vs. decreasing balance' land-use policy for dealing
with hollowed villages. Land Use Policy 29, 11e22.
Long, H., Liu, Y., Li, X., Chen, Y., 2010. Building new countryside in China: a
geographical perspective. Land Use Policy 27, 457e470.
Long, H., Woods, M., 2011. Rural restructuring under globalization in eastern coastal
China: what can be learned from Wales? J. Rural Community Dev. 6, 70e94.
Lu, J., Lora-Wainwright, A., 2014. Historicizing sustainable livelihoods: a pathways
approach to Lead mining in rural Central China. World Dev. 62, 189e200.
Ma, X.L., Heerink, N., Feng, S.Y., Shi, X.P., 2015. Farmland tenure in China: comparing
legal, actual and perceived security. Land Use Policy 42, 293e306.
Mattison, E.H.A., Norris, K., 2005. Bridging the gaps between agricultural policy,
land-use and biodiversity. Trends Ecol. Evol. 20, 610e616.
Menard, S., 1995. Applied Logistic Regression Analysis. SAGE Publications, London,
p. 98.
Meyfroidt, P., 2013. Environmental cognitions, land change, and social-ecological
feedbacks: an overview. J. Land Use Sci. 8, 341e367.
Mullan, K., Grosjean, P., Kontoleon, A., 2011. Land tenure arrangements and rural-
urban migration in China. World Dev. 39, 123e133.
Mushongah, J., Scoones, I., 2012. Livelihood change in rural Zimbabwe over 20
years. J. Dev. Stud. 48, 1241e1257.
Nath, T.K., Inoue, M., 2014. Forest villagers in Northeastern Hill Forests of
Bangladesh: examining their livelihoods, livelihood strategies and Forest Con-
servation Linkages. Small-Scale For. 13, 201e217.
National Bureau of Statistics of China (NBSC), 2012. China Statistical Yearbook.
China Statistics Press, Beijing.
Pichon, F.J., 1997. Settler households and land-use patterns in the Amazon frontier:
farm-level evidence from Ecuador. World Dev. 25, 67e91.
Radel, C., Schmook, B., Chowdhury, R.R., 2010. Agricultural livelihood transition in
the southern Yucatan region: diverging paths and their accompanying land
changes. Reg. Environ. Chang. 10, 205e218.
Scoones, I., 1998. Sustainable Rural Livelihoods: a Framework for Analysis. Working
Paper. Institute of Development Studies, University of Sussex. Institute of
Development Studies, University of Sussex, Brighton, UK, p. 22.
Scoones, I., 2009. Livelihoods perspectives and rural development. J. Peasant Stud.
36, 171e196.
Shackleton, C.M., Shackleton, S.E., Cousins, B., 2001. The role of land-based strate-
gies in rural livelihoods: the contribution of arable production, animal hus-
bandry and natural resource harvesting in communal areas in South Africa. Dev.
South. Afr. 18, 581e604.
Shanghai Statistical Bureau (SSB), 2010. Shandong Statistical Yearbook. China Sta-
tistics Press, Beijing.
Siciliano, G., 2012. Urbanization strategies, rural development and land use changes
in China: a multiple-level integrated assessment. Land Use Policy 29, 165e178.
Soltani, A., Angelsen, A., Eid, T., Naieni, M.S.N., Shamekhi, T., 2012. Poverty, sus-
tainability, and household livelihood strategies in Zagros, Iran. Ecol. Econ. 79,
60e70.
Swets, J.A., 1988. Measuring the accuracy of diagnostic systems. Science 240 (4857),
1285e1293.
Tang, Q., Bennett, S.J., Xu, Y., Li, Y., 2013. Agricultural practices and sustainable
livelihoods: rural transformation within the Loess Plateau, China. Appl. Geogr.
41, 15e23.
Tian, C.H., Li, M.K., 2014. The influence of land rental market on non-farm
employment: evidences from zhejiang, Hubei, Shaanxi province. In: Ifsrap
2013-the First International Forum Studies Rural Areas Peasants, 6, p. 7.
Tittonell, P., Muriuki, A., Shepherd, K.D., Mugendi, D., Kaizzi, K.C., Okeyo, J.,
Verchot, L., Coe, R., Vanlauwe, B., 2010. The diversity of rural livelihoods and
their influence on soil fertility in agricultural systems of East Africa e a typology
of smallholder farms. Agric. Syst. 103, 83e97.
Ulrich, A., Speranza, C.I., Roden, P., Kiteme, B., Wiesmann, U., Nuesser, M., 2012.
Small-scale farming in semi-arid areas: livelihood dynamics between 1997 and
2010 in Laikipia, Kenya. J. Rural. Stud. 28, 241e251.
Urgenson, L., Schmidt, A.H., Combs, J., Harrell, S., Hinckley, T., Yang, Q., Ma, Z., Li, Y.,
Lu, H., MacIver, A., 2014. Traditional livelihoods, conservation and meadow
ecology in Jiuzhaigou National Park, Sichuan, China. Hum. Ecol. 42, 481e491.
Valbuena, D., Verburg, P.H., Bregt, A.K., Ligtenberg, A., 2010. An agent-based
approach to model land-use change at a regional scale. Landsc. Ecol. 25,
185e199.
Van den Berg, M.M., Hengsdijk, H., Wolf, J., Van Ittersum, M.K., Wang, G.H.,
Roetter, R.P., 2007. The impact of increasing farm size and mechanization on
rural income and rice production in Zhejiang province, China. Agric. Syst. 94,
841e850.
Wang, X.X., Shen, Y., 2014. The effect of China's agricultural tax abolition on rural
families' incomes and production. China Econ. Rev. 29, 185e199.
Watson, A., 1983. Agriculture looks for shoes that fit e the production responsibility
system and its Implications. World Dev. 11, 705e730.
Whalley, J., Zhang, S., 2007. A numerical simulation analysis of (Hukou) labour
mobility restrictions in China. J. Dev. Econ. 83, 392e410.
Yan, X.H., Bauer, S., Huo, X.X., 2014. Farm size, land reallocation, and labour
migration in rural China. Popul. Space Place 20, 303e315.
Yi, F.J., Sun, D.Q., Zhou, Y.H., 2015. Grain subsidy, liquidity constraints and food
security-Impact of the grain subsidy program on the grain-sown areas in China.
Food Policy 50, 114e124.
Yu, Q.Y., Wu, W.B., Verburg, P.H., van Vliet, J., Yang, P., Zhou, Q.B., Tang, H.J., 2013.
A survey-based exploration of land-system dynamics in an agricultural region
158 Z. Liu, L. Liu / Journal of Rural Studies 43 (2016) 145e158

of Northeast China. Agric. Syst. 121, 106e116. Zheng, H., Robinson, B.E., Liang, Y.C., Polasky, S., Ma, D.C., Wang, F.C.,
Zhen, N., Fu, B., Lu, Y., Zheng, Z., 2014. Changes of livelihood due to land use shifts: a Ruckelshaus, M., Ouyang, Z.Y., Daily, G.C., 2013. Benefits, costs, and livelihood
case study of Yanchang County in the Loess Plateau of China. Land Use Policy implications of a regional payment for ecosystem service program. Proc. Natl.
40, 28e35. Acad. Sci. U. S. A. 110, 16681e16686.