Spatial-temporal contrasts in integrated urban-rural development in China, 1990-2010.
Li, Yuheng ; Zhang, Zhenghe ; Liu, Yansui 等
infrastructure.
INTRODUCTION
For a long period following the establishment of People's
Republic of China in 1949, urban and rural areas were treated separately
and differently. Since 1978, China has experienced over three decades of
rapid economic growth, with annual growth rates of over 9 per cent.
However, despite this "economic miracle", the country has also
witnessed ever-enlarging urban-rural inequalities across a range of
indicators such as income, education, medical care, provision of
infrastructure and social insurance. For example, while the per capita
urban household income increased from 343 yuan in 1978 to 13,041 yuan in
2007, over the same period the per capita rural household income
increased from 134 yuan to only 3,998 yuan. (1) A cluster of studies
that investigated urban-rural inequalities in China have attributed the
inequalities to factors such as a dualistic urban-rural structure, (2)
urban-biased development strategy, (3,4) market forces (5) and taxation.
(6) Urban-biased policies and related measures like the household
registration system (hukou) were initially formulated on the basis of
the typical socio-economic conditions in China in the period after 1949.
These policies and measures have, however, intentionally diverted
resources (capital, labour and materials) from rural to urban areas, and
induced greater urban-rural inequality in China. (7)
In 2002, the 16th National Congress of the Communist Party of China
(CPC) for the first time stressed the importance of the
countryside's achieving moderate prosperity (xiaokang), and
declared that China's socio-economic development must incorporate
both urban and rural areas alike. This declaration caused a paradigm
shift to the long-standing separation between the urban and the rural,
by placing urban and rural development under the same framework.
Thereafter, a series of policies and measures to achieve integrated
urban-rural development in China was formulated and implemented.
However, China is a vast country of huge socio-economic and geographic
differences--as such, levels of integrated urban-rural development can
be expected to be very different among provinces within different time
periods. (8) This is particularly evident in the different reactions of
the provinces with respect to the transitions of decentralisation,
marketization, urbanisation and globalisation in the post-reform era.
The way in which integrated urban-rural development has evolved in China
has not, to the authors' knowledge, been clearly analysed at the
provincial level. Thus, the aim of this paper was to investigate the
spatial-temporal distribution of levels of integrated urban-rural
development in China in the post-reform period.
The structure of the article is as follows. The first section
provides an understanding of urban-rural linkages and integrated
urban-rural development in China.
The second introduces the research methodology and data, and the
third assesses urban-rural linkages in each province and analyses
spatial-temporal contrasts in integrated urban-rural development. The
conclusion offers a discussion on the research findings.
RESEARCH BASIS AND ANALYTICAL FRAMEWORK
The Evolution of Integrated Urban-Rural Development in China
Generally, the strategy of integrated urban-rural development in
China was first proposed in the 1980s to combat inequalities between
urban and rural areas. However, the local governments charged with
implementing this strategy share two major misunderstandings: first,
that integrated urban-rural development means an integrated distribution
of industries (a mix of agricultural and non-agricultural industries
coexisting in the same area); and second, that integrated urban-rural
development refers to building urban and rural areas alike (i.e.,
transforming villages into cities, and peasants into citizens). (9,10)
The major shortcoming in the interpretation of integrated urban-rural
development lies in the oversight of the unique economic, social and
geographic features of urban and rural areas--that is, the productivity
and other socio-economic factors in cities and rural areas are
different, and cities and rural areas in each province practise
different approaches in the distribution of land, capital, labour and
technology. (11) In fact, the core concept of integrated urban-rural
development regards industry and agriculture, cities and countryside,
and citizens of China and peasants as an integrated whole instead of
isolated parts. (12) According to Chen and Li and Luo and Li, the
fundamental purpose of integrated urban-rural development in China has
been to break down institutional barriers, and achieve coordinated urban
and rural development by strengthening urban-rural linkages. (13,14)
Zhang et al. also point out that integrated urban-rural development
should promote reasonable resource flows (capital, people, materials and
information), agglomeration and profit allocation between urban and
rural areas. (15) Ye lists three main rubrics of integrated urban-rural
development in China: deployment of key factors between urban and rural
areas, supply of primary public goods and services (infrastructure,
compulsory education, health care and social insurance) in urban and
rural areas, and allocation of public resources between urban and rural
areas (this includes the previous two rubrics). (16) In this sense,
integrated urban-rural development can be observed and interpreted as
the consequence of factor flows (labour, capital, goods, information and
technology) and of factor agglomeration between urban and rural areas.
Naturally, an imbalanced distribution of resources may benefit one while
disadvantaging the other: as such, the manner in which urban and rural
areas are linked impacts upon the integration of development between the
two areas.
Integrated urban-rural development in China has been deeply
influenced by government policies. In the early period after the
founding of the PRC, the government emphasised the development of
capital-intensive heavy industries. Against this historical background,
rural areas became the sources of capital, labourers and raw materials
for industrialisation and urban development in China. This economic
strategy has seen the emergence and implementation of many policies and
measures, such as the state-set procurement price for trading of
agricultural products, the hukou system, and agricultural and rural
taxation that disfavoured rural areas. Thus, rural areas in China
gradually lagged behind the cities. Lin and Liu and Lin and Chen point
out that the heavy industry-oriented development strategy had a strong
impact and resulted in higher urban-rural income inequality in China.
(17,18) Luo and Li also argue that urban-biased policies have in fact
limited the direct linkages between cities and the countryside, and
induced urban-rural isolation.
China's reform and opening up since 1978 marked a new
transitional era of integrated urban-rural development. Linkages between
cities and the countryside were strengthened. Cities that were granted
substantial autonomy in decision-making shifted from being passive
agents of the central government to active actors responsible for local
prosperity. The flourishing urban economy provided large market and job
opportunities to the villages and peasants. The household responsibility
system was implemented in rural China in the early 1980s, replacing the
collective farming of the people's commune. The introduction of
this system, which enabled peasants to deal with their surplus, greatly
increased their enthusiasm for agricultural production. The post-reform
era has also seen large-scale rural migration to the cities. Foreign or
Sino-foreign joint enterprises flourished in the manufacturing
industries in eastern China and employed large numbers of rural
labourers, who came mainly from inland provinces such as Guizhou and
Sichuan. According to the China Statistical Yearbook 1999, the
contribution of rural-urban migration to the urbanisation growth
remained at over 60 per cent annually from 1978 to 1989. (19) The large
number of rural migrants, who represented the legacy of cheap labour,
contributed greatly to the urban development of the eastern provinces.
As a strategic way of achieving "urbanisation from
below", township and village enterprises (TVEs) achieved rapid
development in rural China. The strategy--which is to "strictly
control the growth of large cities, rationally develop medium-sized
cities, and vigorously promote the development of small cities and
towns"--encourages peasants to work in TVEs in small towns and
medium-sized cities instead of migrating to large cities. Employment in
the rural non-farm sectors expanded from 9.2 million in 1980 to 191
million in 2004, and the share of rural non-farm sectors in total rural
employment increased from 3 per cent to 38.4 per cent in the same
period. (20)
The early 21st century has seen a historic shift in China from
urban-biased policies to a period of "industry nurturing
agriculture and cities supporting the countryside". The central
government has adopted policies that promote and coordinate harmonious
economic development in both urban and rural areas to achieve overall
well-being in the Chinese society. (21) A slew of measures were taken to
realise this shift, for example rural tax and fees reduction,
agricultural subsidies, support for rural infrastructure construction
and social development. These rural-favoured policies and measures
(which all focused on issues concerning agriculture, villages and
farmers) resulted in the drafting of nine consecutive "No. 1
Central Documents" issued by the Central Committee of the CPC and
the State Council from 2004 to 2012.
Generally, integrated urban-rural development at the provincial
level in China has entered a stage that favours rural areas by
strengthening urban-rural linkages. However, eastern, central and
western regions differ greatly in terms of their level of integrated
urban and rural development. These differences are predominantly due to
socio-economic inequalities between the regions. Following the notion of
"allowing some regions to get rich first and in turn helping other
regions to gradually become rich", the eastern provinces, due to
their socio-economic and geographic advantages, were the first ones to
become rich. In particular, facing competition and opportunities of
globalisation, the eastern provinces received special economic status
from the central government and were subject to preferential policies
(e.g., tax breaks; favourable terms of loan, credit and subsidies; and
higher foreign exchange retention rates). According to the China Foreign
Economic Statistical Yearbook (1980 and 1990), (22,23) China's
actual usage of foreign capital increased from USD109 million in 1979 to
USD3,392 million in 1989. However, over 85 per cent of the foreign
direct investment in the 1980s agglomerated in eastern China. Gradually,
the inequalities between the eastern and inland provinces in China
increased.
Urban-Rural Linkages: The Theoretical Background
Generally, urban and rural areas maintain strong links in the form
of resource flows of people, capital, goods, information and technology.
(24) Potter et al. point out that urban-rural linkages are initiated in
an attempt to take advantage of differentials or complementarities
between urban and rural areas. (25) As both areas differ in productivity
and other factors, such as labour quality and infrastructure, economic
activities in rural areas are primarily agricultural, whereas those in
urban areas are primarily non-agricultural.
To understand how resources flow between urban and rural areas,
central place theory (26) provides the key concept of placing rank order
on the economic activities flowing between villages, towns and cities:
cities are the main suppliers of high-order services like medical
services and education to the surrounding areas, which in turn supply
low-order services like food and other resources to the central place.
The core-periphery model, (27) which was mainly based on the unequal
distribution of power in economy, society and polity, indicates that the
core area is the central realm upon which the surrounding rural
periphery is dependent for the supply of high-order services. The core
area could, as such, evolve into an urban or metropolitan centre with
high potential for innovation and growth while peripheral areas would
experience slow growth or even stagnation, adding to their dependency on
the core area.
The "new economic geography" explains how increasing
returns to scale, agglomeration economies, transport costs and product
differentiation lead to a highly differentiated spatial organisation of
economic activities. Krugman argues that interrelated industry
concentrations, reliable infrastructure, accessibility to the market and
high production returns drive a cumulative process that may result in a
core-periphery economy. The research essentially reveals the
relationship between economic growth and geographic conditions. (28)
Redding and Venables provide evidence that the geography of access to
markets and sources of supply is statistically significant and
quantitatively important in explaining cross-country inequalities in per
capita income. (29) In this sense, resources tend to agglomerate in core
(urban) areas of high profit returns and accessibility to the market,
while the fortunes of rural peripheries are highly reliant upon those
urban areas.
In the field of demographics, rural-urban migration is mainly
considered to be induced by socio-economic factors. In dual economy
theory, Lewis states that surplus rural labourers move from agricultural
sectors to modern industrial sectors due to differences in production
efficiency. (30) Schultz considers migration from lower-productivity
sectors to higher-productivity sectors as a choice made after balancing
the migration cost and potential profit. (31) In economics, rural-urban
interactions between sectors are known as sectoral linkages, which
include rural activities that take place in cities (e.g., urban
agriculture) and activities that are usually classified as urban (e.g.,
manufacturing and services) taking place in the rural areas. (32) These
linkages have developed beyond the traditional division whereby rural
areas are primarily seen as agricultural producers and urban areas are
perceived non-agricultural producers. In most cases, in rural areas,
peasants have been found to undertake non-agricultural jobs that
increase and diversify their income, while in and around cities, growing
urban poverty and the lack of formal employment have catalysed the
development of urban agriculture.
A preliminary assumption of the above analysis is that the level of
integrated urban-rural development of an area (i.e., a province) is
closely related to the flows and agglomeration of resources between
cities and the countryside, and is sensitive to spatial-temporal
dynamics in post-reform China. Thus, we attempted to examine
spatial-temporal contrasts in China's integrated urban-rural
development at the provincial level.
METHODOLOGY AND DATA
Construction of an Urban-Rural Linkage Index
There are 27 provinces and four provincial-level cities (Beijing,
Tianjin, Shanghai and Chongqing) in mainland China (see Figure 1). These
administrative units are grouped into eastern, central and western
regions.
In the above analysis, the integrated urban-rural development at
the provincial level is assumed to be a consequence of flows and
agglomeration of resources between urban and rural areas. In this
connection, the hypothesis is that the level of integrated urban-rural
development in each province varies in response to the presence and
strength of urban-rural linkages. Thus, urban-rural linkages can serve
as a proxy for integrated urban-rural development. The urban-rural
linkage index (URLI) is then constructed to show the level of integrated
urban-rural development of the Chinese provinces. High URLI scores
indicate tight urban-rural linkages and a high level of integrated
urban-rural development.
[FIGURE 1 OMITTED]
Generally, it is not easy to monitor resource flows between urban
and rural areas, since such flows involve complex processes. Lin
identifies the spatial form of urban-rural linkages in the Pearl River
Delta in China by quantifying selected variables of urban-rural
linkages. (33) However, these variables, such as population, employment
and land use intensity, are considered inadequate. Furthermore, the
study relied on data from the year 1991 alone and did not show the
variation of urban-rural linkages over time. Normally, the flows and
agglomeration of resources between cities and the countryside will
induce demographic and economic changes like urbanisation and shifts in
economic structure and income levels, etc. Due to data availability,
five variables are selected to assess the URLI in each province (see
Table 1):
(i) Urbanisation level (the percentage of urban population to the
total population). This variable is selected to show population mobility
between cities and the countryside.
(ii) Ratio of TVE workers to the total rural population. This
variable is chosen to show changes in the employment structure of
secondary and tertiary industries in the countryside. Sectoral linkages
have contributed to the development of nonagricultural industries in
rural areas and employed many rural labourers in off farm industries.
(iii) Ratio of the number of TVEs per thousand people. This
variable describes the development of non-agricultural industries in the
rural areas of each province.
(iv) Ratio of wage income to rural household income. In China,
rural household income includes wage income, household operating income,
asset income and transfer income. Wage income indicates the income
generated by peasants' employment in off-farm industries. This
variable describes the change of rural household income due to sectoral
linkages that have diversified a rural economy and increased rural
household income.
(v) Urban household food consumption. Generally, food consumption
in cities relies heavily on the rural supply. This variable is chosen to
show the material linkage between the countryside and cities.
Considering the possible correlations among these variables,
principal component analysis (PCA) was applied to transform the
variables into a smaller number of uncorrelated variables (named
principal components) that account for most of the variance in the
selected variables. PCA can reduce data dimensionality by introducing a
covariance analysis between factors. The extraction of a few orthogonal
components gives a concise summary of the different variables.
Exploratory Spatial Data Analysis (ESDA)
Upon construction of the URLI, we investigated spatial-temporal
contrasts in levels of integrated urban-rural development at the
provincial level. The central aspect of ESDA is the notion of spatial
autocorrelation or spatial association, the phenomenon in which
similarity of location (observations in spatial proximity) is matched by
value similarity (attribute correlation). (34) Moran's I was mainly
used to test spatial autocorrelation. Global Moran's I measures the
overall clustering and is assessed by examining a null hypothesis.
Rejection of the null hypothesis suggests a spatial pattern or spatial
structure. Local Moran's I examines the spatial autocorrelation and
shows where the clusters or outliers are located and what kinds of
spatial correlation are most important. (35) Global Moran's I is
defined as follows:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)
where [x.sub.i] is the ith observation's value (i = 1, 2 ...
n), n is the number of observations, [w.sub.ij] is a binary system of an
n*n spatial weight matrix. Moran's I ranges from -1 to +1. Positive
values of Moran's I indicate spatial clustering of similar values,
while negative values of Moran's I suggest that high values are
found in the vicinity of low values.
We used a Z score to test the statistical significance that will
determine the acceptance or rejection of the null hypothesis. The
critical Z score values are from -2.58 to +2.58 standard deviations if a
99 per cent confidence level is used. If the Z score is within the
standard deviations (-2.58, +2.58), the research cannot reject the null
hypothesis, implying that overall clustering is very likely to be a
random pattern. If the Z score falls outside the standard deviations
(-2.58, +2.58), then the null hypothesis should be rejected and the
overall clustering displays a significant clustered pattern. The formula
of the Z score is written as:
Z = I - E(I)/[square root of (VAR(I))]
E(I) = -1/n - 1, VAR(I) = [n.sup.2][w.sub.1] + n[w.sub.2] +
3[w.sub.0.sup.2]/[w.sub.0.sup.2](n - 1)(n - 2)(n - 3) - [E.sup.2](I) (2)
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]
where [w.sub.i.] and [w.sub..i] are the sum of the figures in ith
column and ith row respectively.
Thus, local Moran's I can be written as:
Locall = [z.sub.i] [n.summation over (j=1)] [w.sub.ij][z.sub.j] (3)
where [z.sub.i] = [x.sub.i] - [bar.x], [z.sub.j] = [x.sub.j] -
[bar.x] are the deviation of the observed value and the mean value.
The data were taken from the China Statistical Yearbook and China
Village and Township Enterprise Yearbook for the three calendar years
1991, 2001 and 2011. Taiwan, Hong Kong and Macau were not included.
EMPIRICAL RESULTS AND INTERPRETATION
Provincial URLI Scores in the 1990-2010 Period
According to the conventional rule of extracting components that
have eigenvalues greater than one, two components were extracted by PCA
in 1990, accounting for 84.4 per cent of the variance (see Table 2). The
first component, which accounted for 62.9 per cent of the variance, was
the most important component. According to the variable loadings, this
component included variables of urbanisation, the ratio of TVE workers
to the total rural population, the ratio of wage income to rural
household income and the ratio of food consumption to urban household
consumption of nonagricultural products. The second component, which
accounted for 21.4 per cent of the variance, included the ratio of the
number of TVEs per thousand people.
The model for the two principal components in 1990 was formulated
according to the factor loadings of the components and the initial
eigenvalues. The coefficients indicated that the five variables all
contributed positively to the URLI.
F = 0.203[X.sub.1] + 0.452[X.sub.2] + 0.146[X.sub.3] +
0.391[X.sub.4] + 0.342[X.sub.5] (4)
where [X.sub.i] (i = 1,2 ... 5) are the variables; F denotes the
URLI.
The URLI score of each province in 1990 was computed according to
equation (4). The scores were uncorrelated and represented the selected
five variables (Table 3). Similar computations were made for 2000 and
2010. The URLI scores of provinces in these years are listed in Table 4
and Table 5 respectively.
Spatial-temporal contrasts were distinct in the provincial URLI
scores as shown in Figure 2. Generally, provinces of high URLI scores
were located in the eastern parts of China, while the far western
provinces were characterised by low URLI scores. The URLI scores in
central China remained at the medium level. In 1990, provinces of high
URLI scores included provinces in all three regions: those in the
eastern region included Beijing, Tianjin, Liaoning, Jiangsu, Shanghai,
Zhejiang and Guangdong; whereas those in the central and western regions
included Jilin and Guangxi, respectively. There were also provinces in
the three regions that generated lower URLI scores. Sharp contrasts in
URLI scores between provinces were also evident in 2000. Apart from
Hainan, the ten other provinces in eastern China generated high URLI
scores. The URLI scores of the western provinces were low, while those
of the central region stood at a medium level. In 2010, high URLI scores
were found to be concentrated in the southern and southeastern
provinces, and surrounded by inland provinces with medium URLI scores.
Most of the western provinces, as well as Heilongjiang and Henan,
generated low URLI scores. The URLI scores generated imply that
integrated urban-rural development in the eastern provinces remained
high, followed by the inland provinces.
[FIGURE 2 OMITTED]
Spatial-Temporal Contrasts in Integrated Urban-Rural Development
This section examines the degree of clustering in the distribution
pattern of URLI scores and the factors that attributed to
spatial-temporal contrasts in provincial URLI scores. The global
Moran's I in 1990, 2000 and 2010 were 0.303, 0.315 and 0.437
respectively, at the 1 per cent level of significance. This means that
there is less than 1 per cent likelihood that this clustered pattern
could be the result of random chance. Thus, the spatial-temporal
distribution of integrated urban-rural development at the provincial
level was highly clustered in the research period. Moreover, the
clustered pattern is prominent in the research period, as evidenced by
the increases in Moran's I from 0.303 to 0.437.
The LISA (local indicators of spatial association) cluster map was
generated by the GeoDA 095 software programme (see Figure 3). A Moran
scatter plot categorised the nature of spatial autocorrelation into four
types, corresponding to the four different quadrants. Observations with
high values surrounded by observations with high values are in the upper
right quadrant (high-high), while observations with low values
surrounded by those of low values are in the lower-left quadrant
(low-low). Observations with low values surrounded by those of high
values are in the upper-left quadrant (low-high), while the bottom-right
quadrant contains observations with high values surrounded by those of
low values (high-low).
In 1990, over 77 per cent of the provinces fell into the high-high
(26 per cent) and low-low (52 per cent) quadrants, indicating a
polarised clustering pattern of provinces of high or low URLI scores.
Provinces in the high-high quadrant are situated in eastern China, while
provinces in the low-low quadrants are mainly located in central and
western China. In 2010, over 84 per cent of provinces fell into the
high-high (35 per cent) or low-low (49 per cent) quadrants. This change
indicated an increased clustering pattern among provinces of high or low
URLI scores during the research period. This research finding is in line
with the results shown in Figure 3.
Spatial Econometric Model of the Driving Factors
A spatial econometric model was used to analyse the driving factors
behind the spatial-temporal contrasts in China's integrated
urban-rural development at the provincial level. Six variables were
selected to examine the driving factors behind the provincial URLI
scores: per capita GDP (Pgdp), the ratio of non-agricultural production
to total production (Nap); the ratio of exports and imports to total GDP
(Openness), the ratio of people holding a junior college degree or
higher educational qualification to total population (Edu), the length
of highways per square kilometre (Highway) and the location of the
province (L) (where L is the dummy variable indicating an eastern
province if L equals one, and an inland province if L equals zero).
These variables described the economic level, economic structure,
international trade, education level of the population, infrastructure
and location of provinces.
Generally, the spatial error model (SEM) and spatial lag model
(SLM) are two important methods in which spatial interaction is modelled
in spatial regression analysis. The SEM is expressed by equation (5):
Y = X[beta] + [epsilon], [epsilon] = [lambda][W.sub.[epsilon]] +
[mu] (5)
where [beta] is the coefficient of the explanatory variable X;
[epsilon] is the random error term; and [lambda] is the auto-regression
parameter, which measures the spatial dependence. The spatial dependence
means the influence direction and degree from the observed value Y of
the adjacent provinces. W is the spatial weights matrix, [mu] is the
random error term of normal distribution.
[FIGURE 3 OMITTED]
The SLM is expressed in equation (6) as:
y = [rho][W.sub.y] + X[beta] + [epsilon] (6)
where [rho] is the spatial auto-regression coefficient and
[w.sub.y] estimates the spatial correlation degree of the model and
adjusts the influence of other explanatory variables.
Table 6 presents the regression results of the driving factors to
the URLI scores. Generally, economic structure, international trade and
infrastructure were the significant factors with respect to a
province's URLI score. The education level of the population and
location of the province, however, were irrelevant to the provincial
URLI scores. In 1990, only the Nap was found to be positive and
significant at a 5 per cent level. Apart from location and education
level, all the other factors presented positive significance with regard
to the URLI scores in 2000. In 2010, the estimated coefficients of Nap,
Openness and Highway were positive and significant at the 5 per cent
level.
The findings showed that provincial URLI scores are highly related
to the development of non-agricultural industries, international trade
and highway density. With manufacturing as the main pillar of the
economy, China's international trade can help to advance the
development of non-agricultural industries, which will strengthen
cooperation in industrial production between enterprises in the cities
and TVEs in the countryside. Labour mobility, capital and material
linkages between urban and rural areas can also be expected to
strengthen in response to increased production cooperation. Highway
density can contribute to increased accessibility to urban and rural
markets, to rural materials and to the labour force. As shown in Table
7, the average Nap, Openness and Highway were higher in eastern China
than in central and western China (both in 1990 and 2010).
Correspondingly, the URLI scores of provinces in eastern China in 1990
and 2010 were also higher than those of the inland provinces.
CONCLUDING REMARKS
This article provides evidence of spatial-temporal contrasts in the
levels of integrated urban-rural development at the provincial level in
post-reform China. Generally, integrated urban-rural development in
China became spatially clustered over the period of 1990 to 2010, a
pattern that became more evident in the last decade. The eastern
provinces possess a high level of integrated urban-rural development,
followed by the central and western provinces. Such spatial-temporal
contrasts in the level of integrated urban-rural development are
attributed to the development of non-agricultural industries,
international trade and highway density in each province.
The level of integrated urban-rural development was revealed
through an assessment of urban-rural linkages in each province. The
eastern provinces were found to have tighter urban-rural linkages than
the inland provinces, indicating that the development of the urban and
rural areas in the eastern provinces became highly interwoven and
codependent in the post-reform era. Corresponding to the research
assumption, the results also imply that balanced flows and agglomeration
of resources between urban and rural areas in eastern China constituted
resource linkages that shaped these two areas alike. Turning to the
central and western provinces, large urban-rural disparities exist due
to resource flows that predominantly benefit cities and disadvantage the
countryside. Moreover, non-agricultural industries, international trade
and infrastructure in the inland provinces are still lagging behind the
eastern provinces. In particular, the out-migration of educated and
skilled labourers to the cities, especially those in the eastern
provinces, had an impact on the level of integrated urban-rural
development in the inland provinces. Thus, the development of
non-agricultural industries, international trade, and infrastructure
construction in the inland provinces, as well as the coordination of
resource flows between eastern and inland provinces, could contribute to
reducing spatial-temporal contrasts in the levels of integrated
urban-rural development in China.
While the research findings gave proof of the connection between
the level of integrated urban-rural development at the provincial level
and the flows and agglomeration of resources between cities and the
countryside, there is still a lack of theoretical support for integrated
urban-rural development clustering across the provinces.
ACKNOWLEDGEMENTS
This research was supported by two programmes of the National
Natural Science Foundation of China (41130748, 41301190).
Li Yuheng (liyuheng@igsnrr.ac.cn) is Research Associate at the
Institute of Geographic Sciences and Natural Resources Research, Chinese
Academy of Sciences, Beijing, China. He obtained his PhD in Urban and
Regional Studies from the Royal Institute of Technology (KTH), Sweden.
His research interests are urban-rural transformation and integrated
urban and rural development.
Zhang Zhenghe (zhangzhenghe@cau.edu.cn) is Professor of
Agricultural Economics at the College of Economics and Management in
China Agricultural University, Beijing, China. He obtained his PhD in
Agricultural Economics from China Agricultural University. His research
interests include agricultural economics, regional agriculture in China.
Liu Yansui (liuys@igsnrr.ac.cn) is Professor of Land-Use Planning
and Rural Development at the Institute of Geographic Sciences and
Natural Resources Research, Chinese Academy of Sciences, Beijing, China.
He obtained his PhD in Physical Geography from Nanjing Normal
University. His research interests include land-use management and
urbanisation in China.
(1) National Bureau of Statistics of China, Comprehensive
Statistical Data and Materials on 60 Years of New China (Beijing: China
Statistical Press, 2010), p. 23.
(2) Lu Ming and Zhao Chen, "Chengshihua, chengshi qingxiang de
jingji zhengce yu chengxiang shouru chaju" (Urbanisation,
Urban-biased Economic Policies and Urban-rural Inequality), Jingjiyanjiu
(Economic Research Journal) 6 (2004): 50-8.
(3) Cai Fang and Yang Tao, "Chengxiang shouru chaju de
zhengzhi jingjixue" (Political Economics on Urban-rural
Inequality), Zhongguo shehui kexue (Social Sciences in China) 4 (2000):
11-22.
(4) Kanbur Ravi and Zhang Xiaobo, "Fifty Years of Regional
Inequality in China: A Journey through Central Planning, Reform, and
Openness", Review of Development Economics 1 (2005): 87-106.
(5) Gao Qin, "The Social Benefit System in Urban China:
Reforms and Trends from 1988 to 2002", Journal of East Asian
Studies 1 (2006): 31-67.
(6) Tao Ran and Liu Mingxing, "Urban and Rural Household
Taxation in China: Measurement, Comparison and Policy
Implications", Journal of the Asia Pacific Economy 4 (2005):
485-505.
(7) Li Yuheng, "Urban-Rural Interaction in China: Historic
Scenario and Assessment", China Agricultural Economic Review 3
(2011): 335-49.
(8) Liu Yansui, Lu Shasha and Chen Yufu, "Spatio-Temporal
Change of Urban-Rural Equalized Development Patterns in China and its
Driving Factor", Journal of Rural Studies 32 (2013): 320-30.
(9) Feng Lei, "Zhongguo chengxiang yitihua de lilun yu
shijian" (Theory and Practice of Urban-Rural Integration in China),
Zhongguo nongcun jingji (China Rural Economy) 1 (1999): 55-60.
(10) Yuan Zheng, "Zhongguo chengxiang yitihua pingxi ji
gonggong zhengce tanxi" (Appraisal of Urban-Rural Integration in
China and Some Suggestions for Public Policy), Jingji dili (Economic
Geography) 3 (2004): 355-60.
(11) Zhang Hong, Liu Qiuhong and Xiao Yiran, "Jiyu yaosu
liudong de chengxiang yitihua xietiao fazhan silu" (New Ways for
Coordinated Development of Urban-Rural Integration based on Resource
Flows), Jingji tizhi gaige (Reform of Economic System) 6 (2007): 100-3.
(12) Li Yuheng, "Urban-Rural Interaction Patterns and Dynamic
Land Use: Implications for Urban-Rural Integration in China",
Regional Environmental Change 12 (2012): 803-12.
(13) Chen Xiaohong and Li Chenggu, "Chengshihua he Zhongguo
chengxiang yitihua yanjiu" (Study on Urbanisation and Urban-Rural
Integration in China), Chengshiyanjiu (Urban Studies) 2 (2004): 41-5.
(14) Luo Yali and Li Tongsheng, "Zhidu yinsu zai woguo
chengxiang yitihua fazhan guochengzhong de zuoyong fenxi" (Analysis
of the Systematic Function of Urban-Rural Integration in China), Renwen
dili (Human Geography) 4 (2005): 47-51.
(15) Zhang, Liu and Xiao, "Jiyu yaosu liudong de chengxiang
yitihua xietiao fazhan silu" (New Ways of Coordinated Development
of Urban-Rural Integration based on Resource Flows).
(16) Ye Xingqing, "China's Urban-Rural Integration
Policies", Journal of Current Chinese Affairs 4 (2009): 117-43.
(17) Lin Yifu and Liu Mingxing, "Zhongguo jingji de zengzhang
shoulian yu shouru fenpei" (China's Economic Growth
Convergence and Income Distribution), Shijie jingji (World Economy) 8
(2003): 3-14.
(18) Lin Justin Yifu and Chen Binkai, "Urbanization and
Urban-Rural Inequality in China; A New Perspective from the
Government's Development Strategy", Frontiers of Economics in
China 1 (2011): 1-21.
(19) China National Bureau of Statistics, Comprehensive Statistical
Data and Materials on 50 Years of New China (Beijing: China Statistical
Press, 1999), pp. 21-320.
(20) China National Bureau of Statistics, China Statistical
Yearbook 2005, (Beijing: China Statistical Press, 2005), pp. 30-220.
(21) Chen Xiwen, "Review of China's Agricultural and
Rural Development: Policy Changes and Current Issues", China
Agricultural Economic Review 1 (2009): 121-35.
(22) China National Bureau of Statistics, China Foreign Economic
Statistical Yearbook, (Beijing: China Statistics Press, 1980), pp.
21-130.
(23) Ibid., pp. 25-125.
(24) Li Yuheng, "Resource Flows and the Decomposition of
Regional Inequality in Beijing-Tianjin-Hebei Metropolitan Region,
1990-2004", Growth and Change 2 (2012): 335-57.
(25) Robert Potter, Tony Binns, David W. Smith and Jennifer
Elliott, Geographies of Development (London: Pearson Prentice Hall,
2004), pp. 110-358.
(26) W. Christaller, Die zentralen Orte in Suddeutschland (The
Central Places in Southern Germany) (Jena: Verlag von Gustav Fischer,
1933), pp. 30-123.
(27) John Friedmann, Regional Development Policy--A Case Study of
Venezuela (Cambridge: MIT Press, 1966), pp. 34-210.
(28) Paul Krugman, "Increasing Returns and Economic
Geography", Journal of Political Economy 3 (1991): 483-99.
(29) S. Redding and A.J. Venables, "Economic Geography and
International Inequality", Journal of International Economics 1
(2004): 53-82.
(30) Lewis W. Arthur, "Economic Development with Unlimited
Supplies of Labour", The Manchester School 2 (1954): 139-91.
(31) Theodore W. Schultz, "Reflections on Investment in
Man", Journal of Political Economy 5 (1962): 1-8.
(32) Tacoli Cecilia, "Rural-Urban Interactions: A Guide to the
Literature", Environment and Urbanization 1 (1998): 147-65.
(33) Lin George C.S., "Evolving Spatial Form of Urban-Rural
Interaction in the Pearl River Delta, China", Professional
Geographer 1 (2001): 56-70.
(34) L. Anselin, S. Sridharan and S. Gholston, "Using
Exploratory Spatial Data Analysis to Leverage Social Indicator
Databases: The Discovery of Interesting Patterns", Social
Indicators Research 2 (2007): 287-309.
(35) L. Anselin, "Local Indicators of Spatial
Association--LISA", Geographical Analysis 2 (1995): 93-115.
TABLE 1
Description of the Selected Variables
Selected variables Mean Std. Dev
1990 2010 1990 2010
Urbanisation level 0.31 0.49 0.17 0.15
Ratio of TVEs workers to 0.11 0.23 0.07 0.19
total rural population
Ratio of number of TVEs 11.32 15.07 6.96 10.82
per thousand people
Ratio of wage income to 0.09 0.39 0.13 0.13
rural household income
Urban household food 792.48 4637.55 186.03 1074.18
consumption
Selected variables Min Max
1990 2010 1990 2010
Urbanisation level 0.140 0.24 0.77 0.89
Ratio of TVEs workers to 0.002 0.02 0.31 0.91
total rural population
Ratio of number of TVEs 0.250 0.67 24.21 38.97
per thousand people
Ratio of wage income to 0.003 0.12 0.53 0.69
rural household income
Urban household food 540.000 3052.57 1267.00 7776.98
consumption
TABLE 2
Rotated Component Matrix of Analysed Period (1978)
Factor loadings
Selected variables Factor 1 Factor 2
Urbanisation level 0.885 -0.068
Ratio of TVE workers to total rural population 0.908 0.284
Ratio of number of TVEs per thousand people -0.217 0.966
Ratio of wage income to rural household income 0.826 0.178
Urban household food consumption 0.900 -0.149
Initial eigenvalues 3.147 1.072
% of variance 62.935 21.439
TABLE 3
URLI Scores of Provinces in China, 1990
Province Beijing Tianjin Hebei
Score 2.175 1.608 -0.094
Province Jilin Heilongjiang Shanghai
Score -0.804 -1.362 3.833
Province Fujian Jiangxi Shandong
Score 0.839 0.193 0.116
Province Guangdong Guangxi Hainan
Score 1.491 -0.484 -0.826
Province Xizang Shaanxi Gansu
Score -1.358 0.46 -1.061
Province Shanxi Neimenggu Liaoning
Score -0.655 -1.002 0.73
Province Jiangsu Zhejiang Anhui
Score 1.103 1.79 -0.408
Province Henan Hubei Hunan
Score -1.09 -0.466 0.326
Province Sichuan Guizhou Yunnan
Score -0.421 -0.81 -0.831
Province Qinghai Ningxia Xinjiang
Score -1.052 -0.341 -1.533
TABLE 4
URLI Scores of Provinces in China, 2000
Province Beijing Tianjin Hebei Shanxi
Score 2.012 1.816 -0.173 -0.069
Province Heilongjiang Shanghai Jiangsu Zhejiang
Score -0.435 4.402 0.456 0.97
Province Shandong Henan Hubei Hunan
Score 0.161 -0.778 -0.192 0.087
Province Chongqing Sichuan Guizhou Yunnan
Score -0.459 -0.49 -0.948 -0.991
Province Qinghai Ningxia Xinjiang
Score -0.788 -0.312 -1.066
Province Neimenggu Liaoning Jilin
Score 0.077 0.535 -0.006
Province Anhui Fujian Jiangxi
Score -0.585 0.387 -0.371
Province Guangdong Guangxi Hainan
Score 0.607 -0.586 -1.015
Province Xizang Shaanxi Gansu
Score -1.42 -0.162 -0.662
Province
Score
TABLE 5
URLI Scores for China's Provinces, 2010
Province Beijing Tianjin Hebei Shanxi
Score 3.381 1.997 -0.68 -0.09
Province Heilongjiang Shanghai Jiangsu Zhejiang
Score -0.226 1.423 0.624 0.461
Province Shandong Henan Hubei Hunan
Score -0.049 -0.498 -0.213 -0.464
Province Chongqing Sichuan Guizhou Yunnan
Score -1.294 -0.517 -0.524 -0.273
Province Qinghai Ningxia Xinjiang
Score -0.837 -0.598 -0.38
Province Neimenggu Liaoning Jilin
Score -0.474 0.582 2.081
Province Anhui Fujian Jiangxi
Score -0.567 -0.026 -0.775
Province Guangdong Guangxi Hainan
Score 0.795 0.07 -0.933
Province Xizang Shaanxi Gansu
Score -1.237 -0.64 -0.119
Province
Score
TABLE 6
Parameter Estimation of LM Lag and LM Error Models, 1990-2010
Year Cons. Pgdp Nap
1990 Coef. -6.11 *** 9.88 x [10.sup.-6] 5.50 **
Std. Error 2.12 2.12 x [10.sup.-5] 2.56
Log L = -19.42, AIC = 52.84, SC = 62.88
2000 Coef. -3.80 *** 2.41 x [10.sup.-3] *** 3.27 **
Std. Error 1.12 6.21 x [10.sup.-5] 1.46
Log L = -13.51, AIC = 41.02, SC = 51.06
2010 Coef. -6.11 *** 9.88 x [10.sup.-6] 5.50 **
Std. Error 2.12 2.12 x [10.sup.-5] 2.56
Log L = -19.42, AIC = 52.84, SC = 62.88
Year Openness Edu Highway L [R.sup.2]
1990 Coef. 0.05 45.66 0.36 0.14 0.86
Std. Error 0.42 37.84 0.71 0.26
Log L = -19.42, AIC = 52.84, SC = 62.88
2000 Coef. 1.13 ** 6.86 0.55 ** 0.45 0.89
Std. Error 0.46 14.01 0.27 0.29
Log L = -13.51, AIC = 41.02, SC = 51.06
2010 Coef. 1.33 ** 7.60 0.56 ** 0.45 0.86
Std. Error 0.36 15.01 0.27 0.29
Log L = -19.42, AIC = 52.84, SC = 62.88
Notes: LM denotes Lagrange Multiplier; LogL denotes Log likelihood;
AIC denotes Akaike info criterion; SC denotes Schwarz criterion, ***,
** and * indicate the significance levels of 1%, 5% and 10%.
TABLE 7
Variations of Driving Factors in the Three Regions
Region Year URLI Non-agricultural
score industries
Eastern 1990 0.69 0.78
China 2010 1.16 0.92
Annual increase (%) 5.36 1.66
Central 1990 -0.53 0.68
China 2010 -0.09 0.88
Annual increase (%) 18.94 2.57
Western 1990 -0.71 0.66
China 2010 -0.57 0.87
Annual increase (%) 2.14 2.86
Region Year International Highway
trade density
Eastern 1990 0.36 0.34
China 2010 0.69 1.12
Annual increase (%) 6.56 12.62
Central 1990 0.07 0.21
China 2010 0.11 0.90
Annual increase (%) 4.36 15.98
Western 1990 0.07 0.11
China 2010 0.10 0.46
Annual increase (%) 4.17 15.29
