An Empirical Study of the Environmental Kuznets Curve in Sichuan Province, China

The empirical Environmental Kuznets Curve (EKC) literature is colorful but far from conclusive. The environmental Kuznets hypothesis (EKC) confirms an inverse U-shaped relationship between environmental pollution and per capita income. Many authors have analyzed the existence of an EKC for various pollutants. Others have used the EKC framework to identify country characteristics that help to explain the income–environment relationship. But for a local area, such as a province, studies are rare indeed. In this framework, based on the GDP per capita and emissions of industrial waste from 1985 to 2010 in Sichuan Province, China, the relationship is analyzed using regression between economic development and environment in Sichuan Province. Our evidence suggests that there exists a U-shaped or an inverted N-shaped relationship between environmental pollution and economic development in Sichuan Province, that is to say, the environmental Kuznets hypothesis is invalid in Sichuan Province. There are two possible reasons for this conclusion: firstly, KEC curve will not appear at any level of the economic development in Sichuan Province; secondly, the Environmental Kuznets Curve in Sichuan Province exists objectively, but the economic development in Sichuan Province at current stage is not sufficient enough to promote the appearance of KEC curve. However, more attention must be paid to the relation between environmental pollution and per capita income and appropriate environmental policies are required.


Introduction
Through the analysis of economic growth and income gap in 18 countries, American economist Kuznets drew the following conclusions in the study of the relationship between economic growth and income distribution inequality: the inequality of income distribution is rapidly widening at early stages of economic growth, followed by the short-term stability, then it gradually declines at the later stages of growth, while the income inequality in developing countries at the early stages of development is more serious than developed countries, which is called the income Kuznets Curve (Researched by Kuznets, 1995). Suppose that the horizontal axis representing the economic growth and the vertical axis representing the inequality of income distribution, then KC curve is a parabola bent down after upwardly curved, usually referred as the "inverted U-shaped" curve. Based on the income Kuznets Curve theory, scholars Grossman and Krueger empirically analyzed the relationship between environmental quality and per capita income of the North American Free Trade Area for the first time in 1991. They concluded: pollution would increase with the rise in per capita GDP at low income levels, and decline with the growth of per capita GDP at the high-income level (Grossman & Krueger, 1991) .Panayotou first defined the relationship between environment-quality and per capita income as the Environmental Kuznets curve (EKC) using the inverted U-shaped curve in 1993. EKC reveals that the environmental quality would deteriorate with the increase in revenue at first, and then it would improve when income rises to a certain level, namely environmental quality and income show an inverted U-shaped relationship (Panayotou, 1993). Different scholars hold different opinions about the environmental Kuznets curve hypothesis; some support the inverted U-shaped curve, while other studies show that it is U-shaped, N-shaped, monotonously rising and monotonously decreasing. Verbeke and Managi used empirical analysis to test the environmental Kuznets curve, and results showed that the EKC changing trajectory exists in most countries (Verbeke, 2006;Managi, 2006)

Industr
Sichuan in presents th the trend industrial w thousand t meantime, emissions

The Em
The emiss trend, and of an decr emissions where α is constant, β 1 is the time coefficient, β 2 is the quadratic coefficient, β 3 is the cubic coefficient, ε t is the regression error term, AGDP is per capita GDP index in Sichuan Province. For Equations (1) and (2), we assume that: (a) E(ε t ) = 0,(t=1,2,3….,T).
For quadratic curves and cubic curves, different combinations of coefficient symbols have different curve forms.
Model 2 (linear) β 1 > 0, β 2 = β 3 = 0 Linear monotonically increasing Model 3 (linear) β 1 < 0, β 2 = β 3 = 0 linear monotonically decreasing Model 4 (quadratic) β 1 < 0, β 2 > 0, β 3 = 0 U-shaped relationship Model 5 (quadratic) β 1 > 0, β 2 < 0, β 3 = 0 inverted U-shaped relationship Model 6 (cubic) β 1 > 0, β 2 < 0, β 3 > 0 N-type relationship Model 7 (cubic) β 1 < 0, β 2 > 0, β 3 < 0 inverted N-type relationship As shown in Table 1, there are seven models for the curve, and the meanings vary from model to model, broadly speaking, linear monotonically increasing means that the environment quality deteriorates as income increases, linear monotonically decreasing means that the environmental quality improves with income increases. U-shaped relationship means that when income levels are in the lower stages, the environment quality improves as income rises, when the income level is at a high stage; the environment quality deteriorates as incomes rise. What's more, inverted U-shaped relationship means that when income levels are in the lower stages, the environment quality deteriorates as incomes rise and when income levels are in the high stages, the environment quality improves as income rises. Besides, N-type relationship is a kind of curve that as income levels rise gradually, the environment quality deteriorates before further improvement, and finally it falls into deterioration. On the contrary, inverted N-type relationship is totally opposite that as income levels raise gradually, the environment quality first improves before deterioration and at last improves.
www.ccsen  As shown in Table 2, the t statistics of coefficients in the quadratic curve fitting are significant, coefficient of determination in the model is 0.803, the F-statistic is 37.51, indicating that the overall effect of the model is very good; t statistics of coefficients in the cubic curve fitting are also significant, the coefficient of determination in the model is 0.81, F statistics is 27.36, indicating the construction of the model is reasonable, so the quadratic formula for environment and income in Sichuan Province is: The cubic relationship is: As the quadratic curve coefficients β 1 < 0, β 2 > 0, it determined that the environmental quality and income levels in Sichuan Province present "U-shaped" curve in the quadratic curve fitting, as the coefficients of its cubic curves β 1 < 0, β 2 > 0, β 3 < 0, it determined that the environmental quality and income levels in Sichuan Province present "И" type curve in the cubic curve fitting, that is the inverted "N" type curve, as shown in Figure 7 and Figure 8.

Conclusions and Solutions
Through empirical analysis, the relationship between environmental pollution and economic development in Sichuan Province support the environmental Kuznets curve hypothesis, which means the KEC curve does not exist in Sichuan Province. There are two possible reasons for this phenomenon: firstly, KEC curve will not appear at any level of the economic development in Sichuan Province objectively; secondly, the Environmental Kuznets Curve in Sichuan Province establishes objectively, but the economic development in Sichuan Province at current stage is not sufficient enough to promote the inflection point on KEC curve, that is to say the "И" shaped and "U" shaped relationship between environmental pollution and per capita income in Sichuan Province in this study are part of the KEC curve in Sichuan Province in the future.
Economic development provides more financial support for environmental remediation, but the increase in income is not the main incentive of the improvement in the environment quality. The environmental improvement is closely related to the environmental protection awareness of the people in Sichuan, the ideological quality of local people and the relevant policies of the government. Therefore, improvement of the environment should not only be built on the economic development. Sichuan Province must control the industrial emissions, so far industrial emissions is the biggest factor affecting the environment quality in Sichuan, Sichuan needs to phase out some industrial enterprises which are small-scale, high energy consumption, low-income or low production, especially those petrochemical enterprises. Most importantly, we need to promote the transformation of economic structure, shift from the domination of the secondary industry to the domination of the third industry, develop service industry vigorously, and improve people's awareness of environmental protection continuously.
Admittedly, this article has its own limitations and leaves more space for further researches in the future. Due to the lack of research data, it can't demonstrate the existence of Kuznets Curve in Sichuan province accurately. In addition, researchers need to explore a more scientific and effective method when confirm the weight in research.
As for the study of Kuznets Curve in Sichuan province, latter researchers should collect more comprehensive and accurate data, amend the index of environment pollution, and make further discussion about the research methods.