The effects of urbanization on the environment pollution in China ( 2002-2012 )

This paper provides an analytical framework to evaluate whether urbanization could lead to environment pollution or not. This study uses STIRPAT model and recently developed panel regression techniques that allow for heterogeneous slope coefficients and cross-section dependence to model the impact that urbanization has on SO2 emissions for a panel of 28 provinces in China from 2002 to 2012. The estimated contemporaneous coefficients on the urbanization variables are reverse U-shape. The corresponding policy suggestion is that the government should accelerate the urbanization to transfer the economy from a manufacturing based economy to a service based economy.


Introduction
Rapid urbanization in China is having increasing impact on the environment of the growing urban population.More than 450 million of China's 1.3 billion population are now living in urban area where over 70 percent of China's gross domestic product(GDP) are generated, and where the environmental impacts of concentrated human activities are felt the most.Air pollution is one of the most visible environment problems in China's urban life.Especially, the ambient concentrations of sulfate, a fine particulate most damaging to human health, can lead to unsustainable development in China.If urbanization has a significant impact on sulfurs emission(SO 2 ) then this will have implications for sustainable development and urbanization policies.
If urbanization is found to have a positive and statistically significant impact on SO 2 emissions then this can affect forecasting models and urbanization policy.Forecasting models of SO 2 emissions that fail to take into account the impact of urbanization on SO 2 emissions will under forecast sulfur dioxide emissions.Energy and environmental policies that omit the impact of urbanization on SO 2 emissions will likely lead to inaccurate outcomes making sustainable development objectives.If urbanization is found to have a negative and statistically significant impact on SO 2 emissions then this will make sustainable development objectives easier to achieve.And if urbanization is found to have positive and statistically significant impact on SO 2 emissions and then have negative and statistically significant impact on SO 2 emissions, then we will make urbanization develop faster to span the period of negative influence stage.
This paper makes several important contributions to the literature.Firstly, the relationship between urbanization and SO 2 emissions has been studied by a number of authors, but most of this research uses a static model applied to a panel data set.A panel data set is believed to be better than cross-sectional data or time series data by including both time and cross-section dimension.However, static models cannot capture dynamic relationships.This paper uses a static and dynamic framework to model the impact of urbanization on SO 2 emissions in order to compare the results obtained by the two different models.Secondly, previous studies have mostly assumed that the impact of SO 2 emissions is homogeneous across time-series.This is a very strong assumption to make and one that is unlikely to hold across a large grouping of times.In this paper panel regression models are estimated using recent developed techniques that allow for panel unit root test.If panel data exhibits time-series dependence, estimating models with homogeneous slope coefficients may yield estimated coefficients that are biased (Andrew, 2005).In order to solve these problems, this paper use unit root tests and panel cointegration analysis.

A brief review of literature
According to Poumanyvong and Kaneko (2010), the existing literature points to three theories (ecological modernization, urban environmental transition and compact city theories) that are useful for explaining how urbanization can impact the natural environment.
The theory of ecological modernization details how urbanization is a process of social transformation that is an important indicator of modernization.As societies move from low to middle stages of development, environmental problems may increase because in these stages of development, economic growth takes priority (2) In Eq.( 2), counties are denoted by the subscript i (i=1,2...,N) and the subscript t (t=1,2,...,T) denotes the time period.Country specific effects are included through ai and eit represents the random error term.And we use this model to analysis the urbanization on the environment of each provinces in China.Taking natural logarithms of Eq.( 2) provides a convenient linear specification for panel estimation.ln( ) ln( ) ln( ) ln( ) When it comes to estimating Eq.( 4), a distinction can be made between models with homogeneous slope coefficients and models with heterogeneous slope coefficients.If the assumption of homogeneous slope coefficients is made then these models can be estimated using standard panel regression techniques like pooled OLS (POLS) and various fixed effects (FE), random effects (RE), or GMM specifications.Models with heterogeneous slope coefficients can be estimated using mean group (MG) estimators (eg.Pesaran and Smith, 1995;Pesaran, 1997) or variants onmean group estimators.Estimating panel models with heterogeneous slope coefficients is currently an active area of econometrics (eg.Eberhardt et.al., 2013).

Data and resources
The empirical analysis is based on a panel of 29 provinces in China over the period of 2002-2012 because China government had implemented the policy of urbanization reforming from 2002.Due to a lack of data, Tibet is not included and Chongqing is combined to Sichuan.These samples are restricted to those provinces for which the data on the urbanization level denoted as the proportion of the Chinese population living in urban areas(yit), and the data was abstracted from China

Unit root tests
The possible non-stationary property of the data is investigated by applying Bai and Carrion-i-Silvestre(2009) allowing for the structural breaks and cross-sectional dependence.The tests of belong to the socalled "second-generation" of panel unit root tests and have the important advantage to overcome the main limitation of previous tests.Although the tests show that we can reject the null hypothesis that there is a unit root, we can't still confirm that every variable does not have a unit root.Table2 shows that we can find a unit root in lag(3) for SO2, but we can find a unit root for u.So we must use the lag of u to solve the non-stationary problem.

Results
Table 3 presents empirical results for static models with heterogeneous slope coefficients.The estimated coefficient on the urbanization variable is between 0.617 and statistically significant at the 5%.The column(3) in table3 shows that the relationship between urbanization and environment pollution is reverse U shape.That means the environment can damage in the process of urbanization firstly, but as population density in city increase and the public infrastructure can facilitate migrant worker.Hence, The damaging impact of economic growth on the environment reduces by urbanization, and a shift from a manufacturing based economy to a service based economy.The column(4) and column(5) in table3 report the results of the panel cointegration tests by Westerlund and Edgerton(2008), and we can the same conclusions as original panel regression.
Heterogeneous parameter estimates from the dynamic panel model are reported in Table 4. Looking first at the contemporaneous variables, the estimated coefficient on the affluence variable is positive and statistically significant at the 1% level.The values is -0.59 indicating a negative effect.And we can also get reverse U shape adding lagged dependent variable.

Concluding
This paper uses a STIRPAT model to explore the impact that urbanization has on sulfur dioxide emissions in China.It is expected that urbanization will continue rising in China and understanding how urbanization affects SO2 emissions is an important and timely topic to study.A better understanding of how urbanization affects SO2 emissions is necessary from a sustainable development perspective.For static specifications estimated and dynamic specifications estimated, we both get the same conclusion that the relationship between urbanization and SO2 emissions is reverse U-shape.The corresponding policy suggestion is that the government should accelerate the urbanization to transfer the economy from a manufacturing based economy to a service based economy.
DOI: 10.1051/ C Owned by the authors, published by EDP Sciences, 201

Table 1 .
Statistical Yearbook 2003-2013( the National Statistics Bureau of China, 2003-2013).The dependent variable environment pollution index is measured by SO2 emission value(xit), which was also from various issues of China Statistics Yearbook 2003-2013( the National Statistics Bureau of China, 2003-2013).Table 1 presents the definition of the variables and the corresponding summary statistics.variable definitions and descriptive statistics.

Table 2 .
tests of unit root.