INTRODUCTION
In recent decades, the energy consumption growing faster with China’s
rapid economic grows. However, the energy efficiency in China is still at a
low stage, so that how to improve the energy efficiency has become an important
concern. Fossil fuel consumption was increased significantly and resulting in
environmental pollution and climate warming. This has seriously affected the
ecological balance and human health. At present, China has put forward the strategic
goal of improving energy efficiency. Lowcarbon economy requires changing the
present situation of high energy consumption and improving energy efficiency
through the energy structure optimization and innovation. Previous studies showed
that financial development will promote enterprise technological innovation
and industrial upgrading; enterprises can increase the investment in R and D
sector by financial support and this action will finally improve the production
efficiency. However, as China is an emerging country, this financial institution
and financial service is not efficient, so that it still need research on the
effectiveness of how the financial development will effect on China’s lowcarbon
Economic Growth.
With the rapid economic growth in China, China has also facing a sharp energy
consumption growth. China’s total energy consumption equals as 3.62 billion
tons of standard coal in 2012, increased by 3.9% from the previous year. Although
in recent years China’s energy efficiency has improved significantly but
China’s energy efficiency is still at the low stage when compared with
other developed countries. Yuan et al. (2008)
pointed out that China’s economic growth has a stable relationship with
energy consumption. That means China is an energydependent economy and energy
is a limiting factor to output growth in China. By comparing China’s annual
energy consumption and economic growth data from 1990 to 2011, the result indicates
that the energy consumption and economic growth have synchronous growth trend,
as shown in Fig. 1.
In Fig. 1, primary Yaxis (left side) indicates China’s
gross domestic product (GDP), the unit is one hundred million RMB; secondary
Yaxis (right side) indicates energy consumption, the unit is ten thousand tons
of standard coal. From Fig. 1, it shows that the rapid growth
of China’s Gross Domestic Product (GDP) is based on increased energy consumption.
From year 2002 China’s GDP began to rapid growth while energy consumption
has started to increase sharply.

Fig. 1: 
Trend of GDP and energy consumption during 19902011 in China 
Meanwhile, China’s current energy efficiency is low; the economic growth
mainly depends on the energy consumption, so that the extensive economic growth
mode is still the main development pattern.
Osborne and Kiker (2005) pointed out that although
developing countries adopted a variety of measures in climate change mitigation
but the decrease of energy production will effects the economic benefits, so
the final solution will ultimately come from in improving energy efficiency.
Sarkar and Singh (2010) pointed that how to improving
energy efficient is still a challenge in development countries. The result shows
that financial instruments will proved adequate liquidity and can be used for
accelerating the process of lowcarbon economy. Carraro
et al. (2012) pointed that carbon tax revenues are very high in
developing countries, however, the investment in R and D still occupy a low
share of GDP. Therefore, the government should make out effective taxation schemes,
both to reduce greenhouse gas emissions and promote lowcarbon technology innovation.
Chevallier (2011) used factoraugmented vector auto
regression model to analyze how international economic shocks impacts on carbon
markets, the result shows that carbon prices has obvious relationship with most
global economic indicators.
Zou et al. (2011) pointed out that China government
has began to strengthen efforts to propagate and support carbon finance, however,
carbon finance is still faced with management problems and risks. So that, commercial
banks are required to improve their service level and develop new carbon trading
products. YingHua and DanDan (2012) pointed out China’s
lowcarbon economy could be fulfilled through the technical innovation and efficient
energy utilization. There are also some researches about how technical innovation
effects on energy efficiency (Herring and Roy, 2007;
Noailly, 2012; Liu and Shen, 2011).
Based on previous studies, it shows that the development of finance can promote
enterprise technological innovation and industrial upgrading, enterprise through
financial support can strengthen the investment to R and D department and the
enterprises can promote energy efficiency and reduce carbon emissions through
technological innovation.
There are also many academic scholars research about carbon finance and lowCarbon
strategy in China. Li and Colombier (2011) find that
the current China’s energy efficiency standard become to one of the best
practices in the world and with international support such as carbon finance,
the energy efficiency improvement will facilitate city’s transition to
lowcarbon supply in the longer term. Zhou (2010) pointed
out that carbon finance is related to all financial transaction activity that
can reduce carbon emissions. The development of carbon finance can make contribute
to economy transform and accelerate the optimization of economic structure.
Zhao and Zhang (2012) analyzed the internal and external
factors which affect carbon finance in China and put forward the reasonable
path of how to develop carbon finance in China. Also, Zhao
and Zhang (2012) pointed out that China government should set up a full
range of policies and regulations.
Sun and Zhu (2008) using panel data from 23 provinces
in China and find out that the technology innovation level had been gradual
increased in these areas with the financial development. The empirical result
shows that there had been a gradual growth in the TFP and technological progress
from 2001 to 2005 which is tested by variant intercept model in panel data.
Qian and Zhou (2011) analyzed the data of all 28 provinces
in China from 20002008. By using the fixed effects panel data regression with
AR(1) approach to estimate the TFP and the level of financial development in
all regions, the result indicates that after controlling the other relevant
variables, financial development also plays a positive role in technological
progress and industrial upgrading.
To sum up, financial development can promote technology innovation and improve
energy efficiency; at the same time, through the continuous improvement of the
carbon financial market system in China, China’s carbon emissions will
reduce further (Lai et al., 2012; Wang
et al., 2012).
Based on the discussion above, this study first construct an endogenous growth
model and try to find out what is the main factor that effect on lowcarbon
economy growth. Then, using vector autoregression model to make an empirical
analysis about how financial development will effect on energy efficiency.
MATERIALS AND METHODS
Model building: By constructing a closed economy which has five sectors
such as R and D sector, human capital sector, the final product sector, intermediate
products sector and energy production sector. First, assuming that the technology
in R and D sector is nonexclusive; so, technology innovation mainly depends
on the Department’s human capital investment and the existing technical
stock. The R and D sector production function:
A represent the technical stock in economy; δ_{A} represent possibility
of technological innovation, the greater the δ_{A} is the higher
the possibility of technological innovation; H represent the human capital investment.
Human capital mainly depends on the production efficiency of human capital and
input quantity, therefore, the human capital production function:
Also, assuming that the final product production function is a DS function,
so the function of final product sector will be written as:
In equation 1, 0<a_{1}<1; 0<a_{2}<1;
0<a_{3}<1 and _{a1}+a_{2}+a_{3} = 1. H_{r}
represent human capital which has been put into final product sector, Y represent
per capita output, A is the type number of intermediate products, represent
the stock of technical knowledge. Presuming A is continuous; X_{i} is
the quantity of intermediate product i; E is nonrenewable energy that put into
final production department. In the intermediate products sector, the capital
amount can be expressed as:
According to the final product sector production function, the entire intermediate
product X_{i} is asymmetrical and the input requirement is same. For
∀i∈[0, A], so that, X_{i} = X = K/A. Put this into the final
product production function, then the final product production function can
be shown as:
Presuming there is no capital depreciation, therefore, the increase of capital
stock value equals to the total output subtracts consumption and thus the material
capital accumulation equation is:
In energy production sector, assuming S is the stock of nonrenewable; E is
energy into flow the final product production process, the initial stock of
energy is S_{0} and then the stock of energy is:
Based on the derivation of time t, then the consumption changes of nonrenewable
energy can be shown as:
For the consumer preferences, assuming representative consumer has a standard
fixed elastic utility function in the infinite domain as:
In this equation, c represents per capita consumption, L represents the total
population. Assuming the total population is 1, thus, the per capita consumption
is equal to total consumption. In this equation, σ is marginal utility
elasticity in this formula and ρ is the consumer’s time preference.
According to input and output of the final product department, human capital
development of human capital department, technology research and development
of the R and D department, production of product department and nonrenewable
energy consumption, Social dynamic optimization mathematical expression can
be shown as:
By using Hamilton function to calculate the value of this dynamic optimization
model, the equation is:
In the Hamilton function, K, H, A, S are state variables; G, H_{r},
H_{A}, E are control variables; λ_{1}, λ_{2},
λ_{3}, λ_{4} are costate variable and these four costate
variables are the shadow prices of K, H, A, S. According to the optimization
theorem, the firstorder condition for maximize H_{c} is:
λ_{1}, λ_{2}, λ_{3}, λ_{4}
make derivation on t, respectively, then the equation can be shown as:
Transversality conditions for:
On the balanced growth path, each variable growth rates were constant. According
to relationship of consumption, investment and output, it can known that the
economic variables Y, C, K have the same growth rate and g_{r}, g_{c},
g_{x} are the same constant. According to the above formula, g_{r
}and g_{s }can be written as:
Then the only transversality condition is:
(Aσ)g_{r}ρ<0
According to the substance of low carbon economy growth, the necessary condition
to achieve a low carbon economy growth is: g_{r}>0, g_{E}<0,
it means the per capita output increases ceaselessly, the consumption of nonrenewable
energy sources ceaselessly reduce. To achieve this conditions need δ_{H},δ_{A}
sufficiently large; ρ sufficiently small and σ≥1. Therefore, once
the economy has enough human capital accumulation efficiency and technical efficiency,
it will resulting in higher efficiency of technological innovation and can reduced
consumption of nonrenewable energy sources, achieve the lowcarbon economy
growth.
Data collection and processing: Using financial intervnational ratio
to represent the financial development level and using energy efficiency index
to represent the level of Lowcarbon economy. The equation of financial intervnational
ratio is M2/GDP and the equation of energy efficiency index equals GDP/ (energy
consumption). The data was then undertook log processing, noted as LnFIR and
LnEE. All data was collected from “China statistical yearbook 2012”.
RESULTS AND DISCUSSION
ADF unit root test: In statistics and econometrics, an augmented DickeyFuller
test (ADF) is a test for a unit root in a time series sample. By using augmented
DickeyFuller unit root tests, the result as is shown in Table
1.
Through the test results in Table 1, it shows that, LnFIR
and LnEE are nonstationary at 10% critical value. However, after differential
calculation d.LnFER and d.LnCPI are stable, so that VAR model can be used to
analyze the data.
VAR model: Vector Auto Regression (VAR) model is the simultaneous form
of autoregressive model, A VAR (p) model of a time series y (t) has the form:
A_{0}y_{(t)} = A_{1}y_{(t1)}+…+A_{p}y_{(tp)}+ε_{(t)}
According to the analysis above, the VAR regression model of LnFIR and LnEE
can be constructed. Before constructing the VAR model, the lag of VAR Model
should be determined. By using STATA software to calculate the lag length, the
result was shown in Table 2. From the Table
2, the result shows that the optimal lag length is at lag 2. By choosing
lag 2, then the VAR model can be shown as:
LnEE = 0.197+0.568 LnFIR_{t1}+0.198LnFIR_{t2 }+1.807LnEE_{t1}0.989LnEE_{t2}
According to this equation, it shows that the financial development will promote
energy efficiency index increase. LnFIR at lag 1 period increased one percentage
will lead LnEE increased by 0.56% points and LnFIR at lag 2 period increased
one percentage will lead LnEE increased by 0.19% points. so the effect of financial
development to energy efficiency is obvious. Then, by using granger causality
test to analyze the relations between LnFIR and LnEE, the result is shown in
Table 3.
From Table 3, the result shows that LnEE rejected the null
hypothesis as “Excluded LnFIR as the granger reason to LnEE”, so that,
LnFIR is the granger reason to LnEE which means financial development is the
reason promotes energy efficiency index increase. However, it shows that LnEE
is not the reason for LnFIR. At the same time, by taking Johnson cointegration
test to analyze the longterm relations between LnFIR and LnEE, the result is
shown in Table 4. From Table 4, the result
shows that there exist at least one direct cointegration relationship between
LnFIR and LnEE which means that there exists a longterm equilibrium relationship
between financial development and energy efficiency.
Impulseresponse analysis and variance decomposition: Impulseresponse
function and cholesky variance decomposition can be used to further analyze
the VAR model. An impulse response refers to the reaction of any dynamic system
in response to some external change.
The result of Impulseresponse analysis was shown in Fig. 2.
In Fig. 2, Xaxis indicates the time period and Yaxis indicates
the strength of response. From Fig. 2a, the result shows that
when LnFIR received one unit impact, it will lead LnEE increase currently, LnEE
will reach the max at t = 5 period and then begin to stable. It illustrates
there is longterm effect between financial development and energy efficiency
increase. From Fig. 2b, the result shows that when LnEE received
one unit impact, it will lead LnEE increase currently. However, LnEE will reduce
to 0 at =5 period which means the impulse of LnEE only has shortterm effects
to itself. ccording to the impulse analysis results, it shows that financial
development will significant influence energy efficiency.
Table 1: 
Data stationarity test through augmented dickeyfuller analysis 

LnEE is log (energy efficiency) = log (GDP/energy consumption),
LnFIR is log (financial international ratio) = log (M2/GDP), D.LnEE is the
differential of LnEE, D.LnFIR is the differential of LnFIR 
Table 2: 
Result of optimal lagorder selection for VARs 

*Means that lag 2 is the optimal lag order. Abbreviations
are log likelihood (LL), likelihood ratio (LR), final prediction error (FPE),
Akaike’s information criterion (AIC), Schwarz’s bayesian information
criterion (SBIC) and the Hannan and Quinn information criterion (HQIC) 
Table 3: 
Causal relationship test through granger causality analysis 

χ^{2} means chisquared test, Prob>χ^{2}
means the probability that the null hypothesis was established, LnEE is
log (energy efficiency) = log (GDP/energy consumption), LnFIR is log (financial
intervnational ratio) = log (M2/GDP) 
Table 4: 
Longterm equilibrium relationship test through Johnson Cointegration
analysis 

*Means that rank 0 is the optimal rank, that means there exist
at least one cointegration relationship, abbreviation LL is log likelihood 
The variance decomposition indicates the amount of information each variable
contributes to the other variables in the auto regression. It determines how
much of the forecast error variance of each of the variables can be explained
by exogenous shocks to the other variables. The result of variance decomposition
was shown in Fig. 3.
From Fig. 3a, the result shows that the contribution degree
of LnFIR to LnEE is gradually increased, the contribution degree of LnFIR to
LnEE is about 3040% in the all time period which means LnFIR has strong a degree
of contribution to LnEE. From Fig. 3b, the result shows that
the contribution degree of LnEE to LnEE is gradually reduced, the contribution
degree of LnEE to LnEE is 100% at t = 1 period and reduced to 40% at t = 10
period. This shows that financial development has significant effect on the
energy efficiency and can explain the improving the energy efficiency.
To sum up, by constructing the endogenous model, it proved that technological
innovation is the main reason to achieve lowcarbon economic growth, the result
is same to the research from YingHua and DanDan (2012),
Herring and Roy (2007) and Noailly
(2012). The result of empirical study shows that financial development is
the granger reason to the improving of energy efficient. Once financial development
received one unit impact, it will lead energy efficient increase currently.
The result is same to the research from Sarkar and Singh
(2010) which proved financial development will accelerate the process of
lowcarbon economy. However, the empirical study in section 3 did not test whether
financial development will effect on technological innovation, it didn’t
prove that financial development first improve R and D and then the improving
of R and D effects on energy efficient. According to the literature from Sun
and Zhu (2008) and Qian and Zhou (2011), it shows
that financial development in China will promote technological innovation.

Fig. 2(ab): 
Result of impulseresponse analysis for (a) Response of LnEE
when LnFIR received one unit impact and (b) Response of LnEE when LnEE received
one unit impact 

Fig. 3(ab): 
Result of Cholesky variance decomposition for (a) Contribution
degree of LnFIR to LnEE and (b) Contribution degree of LnEE to LnEE 
At the same time, technological innovation will finally effects on the improving
of energy efficient. So that, it can be proved that financial development will
accelerate the process of China’s lowcarbon economy.
CONCLUSION
In conclusions, the result of VAR model shows that financial development is
the Grainger reason to energy efficiency. According to the VAR model, it shows
that the financial development will promote energy efficiency index increase.
Financial development index at lag 1 period increased 1 percentage will lead
energy efficient increased by 0.56% points and financial development index at
lag 2 period increased 1% will lead energy efficient increased by 0.19% points.
At the same time; the result of Johnson Cointegration analysis shows that here
exists a longterm equilibrium relationship between financial development and
energy efficiency. From the variance decomposition results, the result shows
that the contribution degree of financial development to energy efficient is
gradually increased; also the contribution degree of financial development to
energy efficient is about 3040% in the all time period which means financial
development has strong a degree of contribution to energy efficient. The results
prove that financial development can certain explain the improvement of energy
efficiency, so that in the long run, financial development is the main factor
to improve energy efficiency. In recent decades, the demand of energy fuel was
increased dramatically with China’s rapid economic growth, as the financial
industry could be an important support to the lowcarbon economy, so the government
should strengthen the financial factor as reform of the financial system, provides
the corresponding policy support for China’s carbon finance market and
promote innovation of carbon financial products.
In order to improve the effectiveness of low carbon development strategy, the
government should put forward effective policies. First of all, in the process
of low carbon economy development strategy, the government should play the leading
role. Government should give more support and provide a good policy environment
by establishing and improving the relevant policies and regulations. Also the
government should establish the China’s carbon finance system and promote
the innovation of carbon financial products. At the same time, the Chinese government
should ensure the implementation of lowcarbon economy strategy by using tax
and subsidy policies. Second, commercial banks should increase financial products
and services to lowcarbon industry. Because of the money demand for lowcarbon
projects is very large, commercial banks should develop new loan products provide
carbon financial services, so that commercial banks can provide the necessary
financial support to the Lowcarbon innovation in enterprises.
However, this study still has some drawbacks. First, the empirical analysis
did not test the promotion function of financial development for technical innovation;
also the measure of financial development can be more comprehensive. Second,
as China’s energy consumption has obvious regional differences, so that
the impact of financial development on energy efficient will be difference in
different areas. These problems still need further research.