Study on the Impact of Economic Growth and Financial Development on the Environment in China

2.1 The Impact of Economic Growth on the Environment

It is well known that economic growth promotes energy consumption, and energy consumption leads to increasing CO₂ emissions and environmental pollution^[10]. The empirical study of Liu et al.^[11] predicted that over an extending period in the future, CO₂ emissions are going to have a further growing trend with the development of economy. As the researches go deeply, scholars have proposed Environmental Kuznets Curve (EKC) hypothesis. According to their opinions, economic growth and environmental pollution have the relationship of inverted U-shaped, that is, environmental pollution goes up in the early period of economic growth, however, when economic growth achieves a certain level, environmental pollution will get relieved with the further growth of economy^{[12, 13]}. The existing researches have demonstrated that EKC hypothesis is reasonable in most developing countries such as India^{[14, 15]}, Romania^[16], Pakistan^[17] and so on. In addition, Zhao et al.^[3], He et al.^[18] and Han et al.^[19] also found that there is EKC phenomena in China. In the early period of China’s reform and opening up, it is one of the engines for promoting China’s economy to introduce foreign direct investments (FDI). FDI, however, brings a large number of highly polluting industries which do harm to China’s environment without any doubt. Therefore, it can be said that in the few decades China’s economic development is at the expense of the environment. Step by step, through the development of economy, on the one hand, people have recognized the seriousness of environmental degradation; on the other hand, people’s requirements for living standard are becoming higher and higher, especially for the environmental standard. Environmental issues, therefore, are focused on, put an emphasis and getting controlled gradually. On the basis of the analysis above, the following hypothesis is proposed:

H1: Economic growth and environmental pollution in China have the relationship of inverted U-shaped.

2.2 The Impact of Financial Development on the Environment

Financial development is a vital and potential factor in relieving environmental pollution and reducing CO₂ emissions. As Shahba et al.^[8] said that finance sectors are able to influence energy consumption and CO₂ emissions by stimulating the technological progress in the field of energy. Meanwhile, Tamazian et al.^[20] found that financial development can boost better environmental quality through investigating in Brazil, Russia, India, China and other countries.

As the related researches about financial development shows, financial interrelations ratio (FIR) is an important variable to measure the scale and level of financial development. FIR is a proportion of total balance of RMB deposits and loans in financial sectors accounted for gross domestic product (GDP), which means the relevance between finance and economy^[21]. FIR can be used for measuring the scale, level and effectiveness of financial development. Therefore, it is also known as financial intermediation’s scale^[22], the level of financial development^[23], and so on. Jalil et al.^[6] and Tamazian et al.^[20] made empirical researches and concluded that FIR’s rise in China ably contributes to improving the environmental performance.

Another crucial indicator is financial efficiency (FE), which is measured by the proportion of the balance of RMB loans accounted for the balance of RMB deposits in financial sectors. And it is expressed as a kind of efficiency that financial sectors are capable of changing their deposits and savings into loans and investments^[21–23]. Moreover, it reflects the efficiency of supply, operation and allocation about financial resources^[24]. Higher financial efficiency may furtherance greater likelihood of obtaining financing support for high-tech environmental protection enterprises. Therefore, there is no doubt that technology upgrading will be promoted, high energy consumption and high pollution industries will be fallen into disuse, and thus do good to CO₂ emissions reduction and environmental quality rising. It can be seen that the effectiveness and efficiency indicators of the financial development both have played a positive role in reducing CO₂ emissions and environmental pollution. As such, this paper puts forward the following hypotheses:

H2a: Financial interrelations ratio has a negative impact on the environmental pollution in China.

H2b: Financial efficiency has a negative impact on the environmental pollution in China.

2.3 Financial Development’s Moderating Effect Between Economic Growth and the Environment

On the one hand, researchers have demonstrated that the impact of economic growth is likely to vary the degree of environmental pollution in different periods; on the other hand, it is also shown that financial development is instrumental for the environmental protection. However, it is the interaction between economic growth and financial development that may also have a vital impact on relieving the environmental pollution, that is, economic growth and financial interrelations ratio (or financial efficiency) will probably have combined effects on the environment. Specifically, financial interrelations ratio (or financial efficiency) will play a moderating part in the relationship between economic growth and the environment. To be specific, the negative impact of economic growth on the environment is going to be weakened as financial interrelations ratio (or financial efficiency) is heightened. Therefore, this paper proposes the following hypotheses:

H3a: Financial interrelations ratio has a moderating effect on the relationship between economic growth and environment in China.

H3b: Financial efficiency has a moderating effect on the relationship between economic growth and environment in China.

According to the literature review and proposed hypotheses above, this paper presents a theoretical model including the relationship between economic growth, financial development and environmental pollution, just like Figure 1 as follows.

Figure 1

The theoretical model

3.1 Variable Measurement

Environmental pollution is the dependent variable. It is well known that global warming problem mainly stems from CO₂ emissions. Therefore, this paper treats the amount of CO₂ emissions per capita as the measurement for environmental pollution. Firstly, calculate the total amount of CO₂ emission through both all kinds of energy’s CO₂ emissions coefficients estimated by Shen et al.^[2] and total consumption of various energy from “China Energy Statistical Yearbook”. Then, CO₂ emissions per capita can be calculated simply.

Economic growth is the independent variable. GDP per capita is used to represent economic growth in most literatures. Real GDP per capita, however, is a more accurate measurement because of having excluded the influence of price factor.

Financial development is independent and moderating variable. The two aspects of financial development are measured by financial interrelations ratio and financial efficiency. Financial interrelations ratio is RMB deposit and loan balance in financial institutions dividing GDP, financial efficiency is the balance of RMB loans dividing the balance of RMB deposits in financial institutions.

Besides, in this article, the control variables are energy consumption and trade openness. As is known to all that the main sources of CO₂ emissions are energy consumption. Therefore, it is necessary to take energy consumption into the framework, in the meantime avoiding the setting errors to a certain extent. In general, energy consumption is usually expressed as energy consumption per capita. In addition, some scholars also point out that trade openness is an important factor impacting CO₂ emissions^{[5, 7, 9]}. Therefore, simultaneously considering that introducing trade openness can change the closed economy model to an open one, it is indispensable to add trade openness into the analysis. Trade openness is import and export value of goods dividing GDP. It is estimated that more energy consumption and higher degree of trade openness will result in greater economic activities and CO₂ emissions^[7].

3.2 Model Construction

Firstly, on the basis of the impact of economic growth, financial interrelations ratio, financial efficiency, energy consumption and trade openness on the environment in China, Model 1 is constructed.

In order to convince whether EKC hypothesis exists in China, Model 2 is constructed based on Model 1.

For demonstrating whether financial development has a moderating effect on the relationship between economic growth and environmental pollution, Model 3 is established by adding the interaction term of economic growth and financial development on the basis of Model 1.

In order to eliminate heteroscedasticity and keep elastic meaning of regression coefficients, it is needful to convert each variable into natural logarithm. Where t is the time (1978–2012 years), C_t is CO₂ emissions per capita, E_t is energy consumption per capita, O_t is trade openness, G_t is real GDP per capita, Gt2 is squared real GDP per capita, FIR_t is financial interrelations ratio, FE_t is financial efficiency, ln G_t * ln FIR_t is the interaction term of economic growth and financial interrelations ratio, ln G_t * ln FE_t is the interaction term of economic growth and financial efficiency, α_t is intercept term, ε_t is random error term.

3.3 Data Collection

In this paper, the software is Eviews6.0, and the data is collected and calculated through the correlative time series data among 1978–2012 in China. CO₂ emissions are obtained in multiplying all kinds of energy consumption by their CO₂ emissions coefficients. GDP is from “China’s Statistics Yearbook 2013”, and real GDP is got on the basis of the consumer price index in 1978, clearing up the impact of price. Energy consumption comes from “China’s Energy Statistics Yearbook 2013”. Total import and export value of goods also comes from “China’s Statistics Yearbook 2013”. The balance of RMB deposits and loans in financial institutions is from National Bureau of Statistics. Moreover, national population comes from “China’s Statistics Yearbook 2013”. The population in yearbook is counted at the end of the year; however, Zhao et al.^[3] regarded the average population of two years as the population of one year. Thus, we calculate the population like them.

4.1 Unit Root Test

Stationary nature of time series data refers to that their statistical characteristics do not vary over time. Unit root test is a way to judge that. There are three forms in this test: only intercept term, both intercept and trend terms, and neither intercept nor trend term. Therefore, drawing line chart is requisite to determine the form. Besides, lags order is determined by the minimum SIC and SC criteria. Particularly, ADF (Augmented Dickey-Fuller) examination method is used to make unit root test, and the results are shown in Table 1.

In Table 1, testing forms (A, T, K) show intercept term, trend term and lags order; d shows 1st difference; *, ** and *** show significant at 10, 5 and 1 percent level of significance respectively. From Table 1, it is shown that CO₂ emissions per capita, energy consumption per capita, trade openness, real GDP per capita, square of real GDP per capita, financial interrelations ratio, financial efficiency, interaction term of real GDP per capita and financial interrelations ratio, and interaction term of real GDP per capita and financial efficiency exist unit root at 10% significant level. After 1st difference, they are all stationary series. In the opinion of Cointegration Theory, there is long-term equilibrium relationship among the same lags order series. As such, it is suitable for further cointegration test.

4.2 Cointegration Test

The approach of multivariate cointegration is Johansen cointegration test. However, Johansen cointegration test is based on vector autoregressive (VAR) test. Therefore, VAR models are needed to build at first. After identifying the best lags order (4, 3 and 2) of each model, unconstrained VAR models are established. Through testing the VAR models, it is found that they are steady and stationary. Therefore, it is proved that Johansen cointegration test on the basis of VAR models is reliable.

Zhao et al.^[25] pointed out that the lags order of cointegration test can be expressed by the best lags order of unconstrained VAR models minus one. Accordingly, the lags order of cointegration test for Model 1, Model 2 and Model 3 are 3, 2 and 1. Cointegration test for the three models is demonstrated as follows.

1) Cointegration test for Model 1. Its results are shown in Table 2.

In Table 2, * shows to reject the null hypothesis at 5% critical value. The results in Table 2 reveal that, at the 5% significant level, there are six cointegration relationships in Model 1. Therefore, long-term equilibrium relationship is proved among various variables including energy consumption per capita, trade openness, real GDP per capita, financial interrelations ratio, financial efficiency and CO₂ emissions per capita. And moreover, one standardized cointegration equation can be expressed as:

From Equation (4), it is said that coefficient of energy consumption per capita is positive and of greater value, which indicates that since China starts its reform and opening up, the major reason of environmental pollution is plenty of energy consumption, consistent with the existing research and reality. A great deal of energy consumption may be caused by China’s irrational energy structure and inefficient energy usage. Coefficient of trade openness is positive and the value is small, indicating that in the long run there is a positive correlation between trade openness and environmental pollution. Open trade brings quantities of polluting industries from developed countries, which has to lead to “pollution haven” phenomenon in China.

Coefficient of real GDP per capita is positive and of greater value, demonstrating that in the long term there is a positive correlation between economic growth and environmental pollution in China, inconsistent with the phenomenon of “economic growth and carbon emissions decoupling”^[26]. That is to say China’s economic growth remains at expense of the environment. However, improving environmental quality has been more focused on since from recent ten years.

In addition, negative coefficients of financial interrelations ratio and financial efficiency, indicate that in the long term there is a negative relationship between financial interrelations ratio (and financial efficiency) and environmental pollution, which is precisely consistent with the conclusion of foreign scholars. As they said that, financial development will encourage enterprises to use advanced technologies, along with reducing CO₂ emissions and improving environmental performance of domestic products^[10]. According to the empirical analysis of this paper, on condition that other variables don’t vary, financial interrelations ratio increasing by 1% leads up to CO₂ emissions reducing by 0.5148%, and financial efficiency rising by 1% contributes to CO₂ emissions reducing by 0.1199%. It is thus clear that, China’s financial interrelations ratio and financial efficiency will help reduce CO₂ emissions. H2a and H2b, therefore, are supported in the long term. Furthermore, it is also shown that the impact of financial interrelations ratio on CO₂ emissions reduction is stronger than financial efficiency, mainly due to faster expansion of China’s financial scale than financial efficiency in the long run.

2) Cointegration test for Model 2. Its results are indicated in Table 3.

In Table 3, * shows to reject the null hypothesis at 5% critical value. As Table 3 shows that, at the 5% significant level, there are five cointegration relationships in Model 2. Therefore, long-term equilibrium relationship is confirmed among energy consumption per capita, trade openness, real GDP per capita, square of real GDP per capita, financial interrelations ratio, financial efficiency and CO₂ emissions per capita. And one standardized cointegration equation can be written as:

From Equation (5), it is said that in the long term energy consumption per capita and trade openness exacerbate environmental pollution, financial interrelations ratio and financial efficiency alleviate environmental pollution, which is in accordance with the results of Model 1. Coefficient of real GDP per capita is positive, and coefficient of square of real GDP per capita is negative, indicating that in the long term there ever exists inverted U-shaped relationship between economic growth and environmental pollution in China. This is consistent with the opinion of EKC, and meanwhile H1 has been confirmed in the long term. Further, this paper also finds that EKC still exists when bringing financial development into the study framework.

3) Cointegration test for Model 3. Its results are demonstrated in Table 4.

In Table 4, * shows to reject the null hypothesis at 5% critical value. The results in Table 4 show that, at the 5% significant level, there are three cointegration relationships in Model 3. Therefore, long-term equilibrium relationship is as well approved among energy consumption per capita, trade openness, real GDP per capita, financial interrelations ratio, financial efficiency, interaction term of real GDP per capita and financial interrelations ratio, interaction term of real GDP per capita and financial efficiency, and CO₂ emissions per capita. And one standardized cointegration equation can be expressed as:

Above Equation (6) shows that the greater concern should be made about the two interaction terms. Compared with coefficient’s symbols of the two interaction terms in equation (6) as well as coefficients of real GDP per capita in Equations (4) and (6), it can be seen that in the long term China’s financial development have a moderating effect on the relationship between economic growth and environmental pollution. Specifically, with the improvement of China’s financial interrelations ratio and financial efficiency, the degree of environmental pollution from economic growth will be crippled. Therefore, H3a and H3b are supported in the long run. In addition, as shown from coefficients of the interaction terms, the moderating effect of financial interrelations ratio is more forceful than that of financial efficiency. Probably, financial interrelations ratio is the basis of financial efficiency, which may lead to different effects of each other.

4.3 ECM

Johansen cointegration test has indicated the presence of long-term equilibrium relationship among variables, which creates conditions for establishing the error correction model (ECM). ECM on the basis of Model 2 is set up in order to determine the short-term equilibrium relationship among CO₂ emissions per capita, energy consumption per capita, trade openness, real GDP per capita, square of real GDP per capita, financial interrelations ratio, and financial efficiency. ECM is shown in Table 5 as follows.

In Table 5, t value is shown in parentheses, when df = 28; *, ** and *** respectively represents 10%, 5% and 1% significant level passing by t test. As Table 5 demonstrates, coefficient of error correction term is –1.88068, and it passes the test at 1% significant level. Thus, the error terms of the long-term equilibrium make CO₂ emissions per capita fall down in the period of t, in line with the reverse correction mechanism without any doubt.

Coefficients of energy consumption per capita in the first and second lag phases are not significant, indicating that in the short term the impact of energy consumption on CO₂ emissions is not remarkable. Possibly, this reason is that continuous cumulative process characteristics of energy consumption make it impossible to bring out the influence in short-term. In other words, the impact of energy consumption on environmental pollution is still in the incubation period.

Coefficients of trade openness in the first and second lag phases are negative and significant. As it shows that in the short term trade liberalization can help to alleviate environmental pollution because of more environmentally friendly and cleaning products imported.

From positive and significant coefficients of real GDP per capita in the first and second lag phases, as well as negative and significant coefficients of square of real GDP per capita in the first and second lag phases, it is known that in the short term China’s economic growth is still a very important factor for environmental pollution. And moreover, it is also proved that in short-term China still exists EKC, consistent with the opinion of Zhao et al.^[3] and Pao et al.^[27]. Therefore, H1 is also confirmed in short-term.

Coefficients of financial interrelations ratio and financial efficiency in the first lag phase are non-significant, however, in the second lag phase they are significant. That indicates China’s financial development for alleviating environmental pollution is behind-time in the short term. On the one hand, if environmental and high-tech industries want to get financial support, they have to go through a qualification examination procedure; on the other hand, they have inherent cycle of production activities. Therefore, it is reasonable to be falling-behind for the impact of financial development on the environment. Besides, their coefficients also demonstrate that the effect of financial efficiency is stronger than financial interrelations ratio in the short term, which is not consistent with the long-term situation. Mainly, it is well-known that financial efficiency is improved faster and more flexible in a short period.