## Abstract

In this paper we examine the impacts of carbon tax policy on CO2 mitigation effects and economic growth in China by using a dynamic energy-environment-economy computable general equilibrium (CGE) model. The results show that 30, 60, and 90 RMB per ton CO2 of carbon tax rate will lead to a reduction of CO2 emissions by 4.52%, 8.59%, and 12.26%, as well as a decline in the GDP by 0.11%, 0.25%, and 0.39% in 2020, respectively, if carbon tax revenues are collected by the government. Moreover, with energy efficiency improvements the CO2 emission per unit of GDP will equally drop by 34.79%, 37.49%, and 39.92% in 2020, respectively. Negative impacts on sectors and households will be alleviated if carbon tax revenues are returned to these sectors and households.

## Keywords

dynamic energy-environment-economy CGE model ; carbon tax ; mitigation effect ; economic impact

## 1. Introduction

The Chinese government has announced to lower its CO2 emission per unit of GDP by 40%–45% by 2020 compared with the 2005 level, and to increase the share of non-fossil energy in primary energy consumption to 15% by 2020. China is currently in the period of accelerated industrialization and urbanization, with steadily rising energy demand, which means that China’s CO2 emissions will continue to increase in the short term. China is under high pressure and faces difficulties in controlling the growth of CO2 emissions.

China has primarily been relying on administrative measures to promote energy saving and CO2 mitigation. However, market-based policies mainly aim at influencing the market factors, i.e., commodity supply and demand, price, and competition, to stimulate economic agent to reduce energy consumption and CO2 emissions.

Carbon tax, as a market-based approach, is a highly praised mitigation method. Firstly, it leads enterprises to control CO2 emission through price change. Secondly, the implementation cost of carbon tax is relatively lower because it can be carried out on the basis of existing tax. Thirdly, it can create a good reputation and position for climate change negotiation. However, some disadvantages also arise from carbon tax. On the one hand, imposing carbon tax may bring negative impacts on economic growth and household income, which is likely to meet resistance from the related interest groups. On the other hand, China may face energy supply shortages in the long term, while the energy demand in China appears rigid. The increased cost owing to carbon tax may be transferred to the energy demand side, which will probably alter CO2 mitigation effects.

The computable general equilibrium (CGE) model, which is used for simulating and analyzing policy effects, has recently become a popular tool for policy researchers. Carbon tax is turning into one of the research foci in the field of CO2 mitigation policies. The first type of studies which concentrated on carbon tax by using the CGE model is to simulate the required carbon tax rate under a constraint of mitigation targets [ Zhang, 1998  ; Garbaccio et al., 1999  ; He et al., 2002  ; Wang et al., 2005  ; Liang et al., 2007 ]. The second type involves mitigation and economic effects under different carbon tax policies [ Cao, 2007  ;  Cao, 2009 ; Wang et al., 2009  ; Su et al., 2009a ]. According to Zhang [1998], the negative impact of carbon tax on GDP would be reduced if the carbon tax revenues were used to offset reductions in indirect taxes. Double dividend of environmental tax has also been discussed [ Bovenberg and de Mooij, 1994 ; Bovenberg and van der Ploeg, 1994  ;  Parry, 1995 ]. In regard to China’s carbon tax design, Cao [2009] has simulated the impacts of carbon tax policies with different tax rates and different use of carbon tax revenue. However, CO2 emission per GDP is predicted to decrease quickly, with an average annual rate of nearly 4% in the future.

In this study, we indicate that additional efforts are needed to achieve the target of reducing CO2 emissions by 40%–45% in 2020 compared with 2005 levels. In this research, we set a more reasonable baseline scenario, which is based on the results of a bottom up model and actual energy efficiency improvement, and then comprehensively assess the mitigation and economic effects of carbon tax policies, as well as the contribution to carbon intensity target for 2020.

## 2. Analysis of China’s CO2 intensity target for 2020

Predictions of GDP from 2011 to 2020 are based on China’s 12th Five-Year Plan, as well as on simulation by Pan et al. [2001] , He [2001] , AGDRCSC [2005] , Wei et al. [2008] , and EIA [2009] . Here, three scenarios with different GDP growth rates are taken into account (Table 1 ). Based on the reasoning that CO2 accounts for almost 80% of the greenhouse gases (GHGs), and that around 90% of CO2 emissions come from the burning of fossil fuels, in this paper we only consider the energy related CO2 emission and proportionately reflect the 2020 carbon intensity target. Energy related CO2 emissions and reduction under three economic development scenarios in 2020 are displayed in Table 2 .

Table 1. GDP growth rates under three economic development scenarios (%)
Scenario 2008 2009 2010 2011–2015 2016–2020
High economic growth 9.6 9.1 10.3 9.0 8.0
Middle economic growth 9.6 9.1 10.3 7.5 6.5
Low economic growth 9.6 9.1 10.3 6.0 5.0

Table 2. CO2 emission and reduction under three economic development scenarios in 2020 (Gt)
Scenario Consistent with 2005 Reduction by 40% Reduction by 45%
Total emission High economic growth 19.41 11.65 10.68
Middle economic growth 16.89 10.13 9.29
Low economic growth 14.66 8.80 8.07
Reduction High economic growth 7.76 8.73
Middle economic growth 6.76 7.60
Low economic growth 5.87 6.60

The USA announced a 17% CO2 emission reduction target by 2020 from 2005 levels. The EU and its member states are committed to an independent quantified economy wide emission reduction target of 20% by 2020, compared with 1990 levels. This target could be increased to 30% under the conditions set out by the European Council. It should be noticed that total CO2 reduction of the USA and EU are about 2.5–3.2 Gt according to the commitment, which is less than half of the CO2 reduction under middle economic growth scenarios for China. While the conditional pledge of Japan for 2020 is even less than 1/10 of CO2 reduction estimated for China. It gets clear that it is not an easy task to achieve the 2020 carbon intensity target set for China. However, China is bearing the high international pressure as it will be still the world’s leading CO2 emitter, even when reaching the carbon reduction target for 2020. The total CO2 emissions will account for approximately 1/4 of global emissions in 2020.

## 3. Methodology and data

### 3.1. Methodology

The dynamic energy-environment-economy CGE model used in this paper covers 39 sectors, including 1 agriculture sector, 36 industry sectors, and 2 service sectors, and relates to two sets of households and three production factors (labor force, capital, and energy). The inputs of energy factor are constituted of nine energy sectors, that is, coal, oil, natural gas, oil refined products, coke, fuel gas, thermal power electricity, other electricity, and heat.

The basic model is built according to the structure by Lofgren et al. [2002] , while the energy is embedded into the production module according to Wu and Xuan [2002] . Since the elasticity of substitution varies among different inputs, the production is described by a multi-level nested structure. Substitution occurs firstly among different energies, then between capital and energy, and finally between a combination of capital-energy and labor. The constant elasticity of substitution (CES) function (Eq. (1)) is adopted to describe the substitution relationship among them. Intermediate commodities enter into the model with a Leontief structure (Eq. (2)) to reflect the assumption of no substitution among different intermediate commodities, as well as between intermediate commodities bundles and factor bundles, which consists of a labor and a capital-energy bundle. The main production structure is presented in Figure 1 .

 Figure 1. Structure of the production module in the dynamic CGE model

CES function:

 ${\displaystyle Y_{i}=f\left(X_{i1},X_{i2}\right)={\left(\alpha X_{i1}^{\rho }+\beta X_{i2}^{\rho }\right)}^{1/\rho }}$ ${\displaystyle =A{\left(\alpha X_{i1}^{\rho }+\left(1-\alpha \right)\beta X_{i2}^{\rho }\right)}^{1/\rho }}$
( 1)

Leontief function:

 ${\displaystyle Y_{i}=min\left(\alpha X_{i1},\beta X_{i2}\right)}$
( 2)

Where, Yi denotes output of sector i ; Ai denotes shift parameter of sector i, α denotes share parameter of input Xi 1 , β denotes share parameter of input Xi 2 , ρ denotes substitution parameter.

Economic growth is driven by endogenous investment, consumption and net export. Investment is determined by savings which come from household, government, enterprise and abroad. Household income is used for tax payment, consumption and saving, while government income is expended for transfer payment, subsidy, consumption and saving. Household and government consumption are co-determined by income and consumption preference, and described by the extended linear expenditure system. The Armington assumption is used to distinguish identical domestic goods and imported (exported) goods. World price is exogenous and China is treated as price-taker in all markets. The year 2007 is regarded as the base year in this simulation. The dynamical modeling is realized by capital accumulation, labor growth and technological improvement.

### 3.2. Data

The basic dataset used in this model is the China 2007 Social Accounting Matrix, which is constructed by using 135-sectors input-output table [ NBSC, 2009 ] and other data of 2007 including custom, tax, international balance of payment, and the flow of funds. The 135-sectors input-output table is aggregated and disaggregated into 39 sectors. Substitution elasticity among different energy, energy and capital, energy-capital combination and labor refers to the researches of Wu and Xuan [2002] , and Paltsev et al. [2005] . Substitution elasticity between import and domestic commodities refers to GTAP-6 . The data of labor by sectors come from The Fifth Population Census and the China Economic Census Yearbook 2004 [ LGSCFNEC, 2004 ]. The data of fixed assets investment and population are from the China Statistical Yearbook 2008 [ NBSC, 2008 ].

Energy consumption should be treated in physical terms, while the input-output table provides energy consumption in each agent in monetary terms. The base year conversion coefficient is obtained by converting monetary terms to physical terms [ NBSC and CDNEA, 2008 ], and is also used in the forecast of future energy consumption with the CGE model. As recommended by the IPCC [2006] , the CO2 emissions are calculated by the method of multiplying each type of fossil fuel consumption with its CO2 emission factors and oxidation rates.

## 4. Scenario design

### 4.1. Baseline

The GDP growth rates for 2008–2010 are actual measured data, while from 2011–2020 are estimated according to the middle economic growth rate mentioned in Table 1 .

It is assumed that the proportion of effective labor force to total population is constant in the simulation period, while the predicted population is taken from the Research Report on National Population Development Strategy [ NPDSR, 2007 ].

The commodity price index (CPI) is based on historical data. The average value of the CPI from 1990–2010 is 102.4 (once outliers of 114.7, 124.1, and 117.1 for 1993, 1994, and 1995 are eliminated). According to international experience, it is likely that the CPI will not always be at such a high level, therefore the CPI is anticipated to be 102 per year from 2011 to 2015, and 101 per year from 2016 to 2020.

It is assumed that energy efficiency improvements exist under the baseline scenario, which refers to the AIM model with calibrations according to the Medium and Long Term Energy Conservation Plan including the efficiency gap of developed countries. The average annual increase rate of energy efficiency is 2.1%, and for details refer to Shi and Zhou [2010] .

As coal prices are primarily controlled by the domestic and central government, it is endogenous by the equilibrium of market demand and supply. Although oil and gas prices are guided by the government, the external dependence of oil is as high as 50%, and the price reformation is gradually reducing the gap of the international market. Hence, the tendencies of domestic oil and gas consumption prices are exogenous and in line with international markets, which refer to the forecast by the World Energy Outlook 2009 [ IEA, 2009 ] and the Poles model [ Kitous et al., 2010 ]. Electricity prices for China are provided by the predictions of Kitous et al. [2010].

Global prices of commodities are taken from the predictions by Peterson et al. [2011] for the baseline run. The future appreciation of the RMB is also considered and assumed to rise by 10% until 2020.

### 4.2. Carbon tax scenarios

Economic and environmental effects compared with baseline are analyzed in carbon tax scenarios with the following features (Table 3 ). Scope on carbon tax. It is suggested that carbon tax should be levied at the beginning only on CO2 emissions from fossil fuel. This is based on the reason that CO2 accounts for almost 80% of GHGs and 90% of CO2 emissions come from fossil fuels, moreover, these emissions are relatively concentrated and easy to measure.

Table 3. Carbon tax revenue scenarios
Scenario Tax rate (RMB per ton CO2 ) Carbon revenue
S1a 30 Government income
S2a 60 Government income
S3a 90 Government income
S2b 60 50% as government income, 50% reduction of other tax rates of the most impacted sectors
S2c 60 50% as government income, 50% as household income
S2d 60 50% as household income, 50% reduction of other tax rates of the most impacted sectors

Carbon tax incidence: consumer or producer? Although carbon tax levied on the consumption process is good for restraints on energy demand, it is easier for tax collection and administration, as well as source control when carbon tax is imposed on the production process.

Date for carbon tax implementation. The first commitment period of the Kyoto Protocol is until the end of 2012, and developing countries should take measureable, reportable, and verifiable (MRV) actions against climate change according to the “Bali Road Map”. In China, carbon tax should be coordinated with other energy related taxes, while the period of tax reform can be regarded as a good opportunity to include carbon tax. For the above reasons, the carbon tax is suggested to start from 2013, and being constant until 2020.

Carbon tax rate. The carbon tax rate primarily ranges from 7 to 44 Euro (73 to 458 RMB) per ton CO2 in international experiences [ Su et al., 2009b ]. The tax rate should not be too high in the beginning, considering the negative impacts on Chinese enterprises and international competitiveness. Average clean development mechanism (CDM) price in international market is from 3.1 Euro (32 RMB) in 2004 to 14.8 Euro (151 RMB) in 2008 per ton CO2 . Hence, different tax rates of 30, 60, and 90 RMB per ton CO26 are set for the simulation.

Carbon tax revenue use. Three types of carbon tax revenue use are considered in this paper. (a) Treatment of the carbon tax revenue as income of the central government. (b) Imposing carbon tax on energy producer while cutting other taxes off in most impacted sectors, proportionately. (c) Transfer of the carbon tax revenues to the households (Table 3 ).

## 5. Results

### 5.1. Baseline scenario

The baseline scenario shows that the GDP will reach 70.46 trillion RMB in 2020, with the primary, secondary and tertiary industries accounting for 8.03%, 47.98% and 43.99%, respectively. China’s total energy consumption will grow from 2.53 Gtce in 2007 to 5.66 Gtce in 2020, which means that a GDP growth rate of 82.97% can be achieved while only 61.20% more energy is consumed. Total energy related CO2 emission is expected to reach 11.52 Gt in 2020, with a growth rate of 5.86% per year from 2007 to 2020, which corresponds to a 31.79% reduction of carbon intensity compared with 2005 (Table 4 ).

Table 4. Simulation results of major indicators under the baseline scenario
Year GDP (trillion RMB) Energy consumption (Gtce) Energy related CO2 emission (Gt)
2007 27.16 2.53 5.50
2009 32.48 3.01 6.48
2011 38.51 3.51 7.51
2013 44.50 3.97 8.40
2015 51.43 4.48 9.36
2018 62.12 5.16 10.61
2020 70.46 5.66 11.52

### 5.2. Results of different carbon tax rate scenarios

CO2 emissions and energy consumption are restricted by carbon tax policies. Total CO2 emissions are predicted to be 11.00, 10.53, and 10.11 Gt under carbon tax rates of 30, 60, and 90 RMB per ton CO2 in 2020, respectively. This is 4.52%, 8.59%, and 12.26% lower than in the baseline scenario, and compared with 2005 reduces the carbon intensity by 34.79%, 37.49%, and 39.92% in 2020, respectively. Total energy consumption will grow from 2.53 Gtce to 5.47, 5.29, and 5.14 Gtce in 2020, which is 3.43%, 6.51%, and 9.28% less than in the baseline scenario under a carbon tax rate of 30, 60, and 90 RMB per ton CO2 , respectively (Table 5 ).

Table 5. CO2 emission and energy consumption for different carbon tax rate scenarios in 2020
Scenario CO2 emission (Gt) CO2 reduction rate compared with the baseline scenario (%) Decreases in CO2 intensity rate from 2005 (%) Total energy consumption (Gtce) Energy saving rate compared with the baseline scenario (%)
S1a 11.00 –4.52 34.79 5.47 –3.43
S2a 10.53 –8.59 37.49 5.29 –6.51
S3a 10.11 –12.26 39.92 5.14 –9.28

Table 6 displays the economic effects under different carbon tax rate scenarios in 2020 compared with the baseline scenario. The GDP is modestly negative impacted, with a change rate of –0.11% to –0.39% under carbon tax rate of 30 to 90 RMB per ton CO2 . The government consumption will increase due to a larger government income from carbon tax revenues. The carbon tax policy will have adverse effects on household income and consumption as these sectors are less competitive in international market, since carbon tax will lead to higher energy and energy intensive commodities prices. It is obviously that the total export will drop significantly with carbon tax policy compared with the baseline scenario.

Table 6. Economic effects in different carbon tax rate scenarios versus the baseline scenario (%)
Scenario Year GDP Government consumption Household consumption Net export Government income Household income
Urban Rural
S1a 2013 –0.11 1.52 –0.45 –4.90 1.21 –0.53 –0.49
2020 –0.11 1.24 –0.40 –5.66 0.94 –0.51 –0.51
S2a 2013 –0.24 2.90 –0.88 –10.11 2.27 –1.05 –0.98
2020 –0.25 2.36 –0.77 –11.55 1.77 –0.99 –0.98
S3a 2013 –0.39 4.13 –1.29 –15.56 3.20 –1.55 –1.45
2020 –0.39 3.38 –1.12 –17.62 2.50 –1.45 –1.43

The energy production sectors will be most impacted, especially the coal mining and washing sector, whose output will reduce by 8.92%, 16.74%, and 23.59% in 2020 under carbon tax rate of 30, 60, and 90 RMB per ton CO2 , compared with the baseline scenario, respectively. The energy intensive sectors will also be negatively affected, but not so dramatically (Table 7 ).

Table 7. Change rate of output in different carbon tax rate scenarios compared with the baseline scenario in 2020 (%)
Industry S1a S2a S3a
Coal mining and washing –8.92 –16.74 –23.59
Coking –4.66 –9.36 –14.01
Raw chemical materials and chemical products –0.76 –1.51 –2.22
Smelting and pressing of ferrous metals –0.63 –1.24 –1.84
Ferrous metals mining and dressing –0.62 –1.23 –1.82
Glass and glass products manufacturing –0.54 –1.07 –1.58
Other nonmetal mineral products –0.50 –0.98 –1.45
Gas production and supply –0.50 –1.00 –1.51
Metal products –0.48 –0.95 –1.41
Nonferrous metals mining and dressing –0.48 –0.95 –1.42

Note: This table outlines the top 10 negative impacted sectors, and sectors are sized down according to influence extent

### 5.3. Results of different scenarios of carbon tax revenue

The use of carbon tax revenues will have different impacts on the CO2 emission and the economy. Table 8 outlines the CO2 emission and energy consumption under different carbon tax revenue scenarios compared with the baseline scenario in 2020. Total CO2 emissions reach 10.53, 10.61, 10.53, and 10.63 Gt in 2020 under scenarios of s2a, s2b, s2c, and s2d, respectively, which are 8.59%, 7.92%, 8.59%, and 7.72% less than in the baseline scenario. This simultaneously decreases the carbon intensity by 37.49%, 37.06%, 37.45%, and 36.89% from the 2005 level. Total energy consumption is predicted to be approximately 5.3 Gtce in different carbon tax revenue scenarios in 2020. It can be seen that using carbon tax revenue as household income has similar CO2 mitigation effects as government income, while levying carbon tax when reducing other tax rates of most impacted sectors will have relative smaller CO2 mitigation effects.

Table 8. CO2 emission and energy consumption in different carbon tax revenue scenarios compared with the baseline scenario in 2020
Scenario CO2 emission (Gt) CO2 reduction rate compared with baseline (%) CO2 intensity reduction compared with 2005 (%) Total energy consumption (Gtce) Energy saving rate compared with baseline (%)
S2a 10.53 –8.59 37.49 5.29 –6.51
S2b 10.61 –7.92 37.06 5.32 –5.97
S2c 10.53 –8.59 37.45 5.30 –6.49
S2d 10.63 –7.72 36.89 5.33 –5.79

In Table 9 the economic effects in different carbon tax revenue scenarios compared with the baseline in 2013 and 2020 are displayed. It is illustrated that using carbon tax revenue as household income will have a negative effect on the GDP, whereas the adverse impacts on household income and household consumption can be alleviated. Although the CO2 mitigation effects are less obvious with the policy of levying a carbon tax while reducing other tax rates of more vulnerable sectors, the GDP loss appears to be smaller and the negative impacts on net export will be alleviated to a certain extent.

Table 9. Economic effects in different carbon tax revenue scenarios compared with the baseline scenario (%)
Scenario Year GDP Government consumption Household consumption Net export Government income Household Urban income Rural
S2a 2013 –0.24 2.90 –0.88 –10.11 2.27 –1.05 –0.98
2020 –0.25 2.36 –0.77 –11.55 1.77 –0.99 –0.98
S2b 2013 –0.19 2.05 –0.65 –7.39 1.51 –0.82 –0.86
2020 –0.20 1.80 –0.62 –8.98 1.33 –0.83 –0.89
S2c 2013 –0.29 1.45 –0.33 –12.17 0.96 –0.69 0.19
2020 –0.31 1.18 –0.36 –13.73 0.74 –0.72 –0.18
S2d 2013 –0.23 0.39 –0.04 –8.88 0.00 –0.41 0.35
2020 –0.26 0.44 –0.17 –10.47 0.00 –0.51 –0.06

It can be seen from Table 10 that the change rate of output under the policy of using carbon tax revenue as government income and household income are similar, while reducing other taxes of those most impacted sectors will decrease the negative impacts to some degree.

Table 10. Change rate of output in different carbon tax revenue scenarios compared with the baseline scenario in 2020 (%)
Industry S2a S2b S2c S2d
Coal mining and washing –16.74 –15.72 –16.74 –15.45
Coking –9.36 –8.46 –9.38 –8.23
Raw chemical materials and chemical products –1.51 –1.28 –1.54 –1.25
Smelting and pressing of ferrous metals –1.24 –1.04 –1.27 –1.02
Ferrous metals mining and dressing –1.23 –0.68 –1.27 –0.57
Glass and glass products manufacturing –1.07 –0.93 –1.08 –0.90
Gas production and supply –1.00 –0.91 –0.96 –0.85
Other nonmetal mineral products –0.98 –0.86 –1.01 –0.85
Metal products –0.95 –0.80 –0.99 –0.79
Nonferrous metals mining and dressing –0.95 –0.77 –1.03 –0.80

Note: This table outlines the top 10 negative impacted sectors, and sectors are sized down according to influence extent

## 6. Sensitivity analyses

To get a reliable CO2 emission projection and evaluate the economic and CO2 mitigation effects the baseline scenario is investigated. Hence, it is worth to test the robustness of the baseline results in face of the uncertainties of future projections. To perform a sensitivity analysis of the baseline results, several key parameters are examined and divided into three types of alternative scenarios. The first type is a high or low energy efficiency improvement, since energy efficiency improvement makes great contribution to the reduction of carbon intensity. The second category is chosen to understand the robustness of model by changing the elasticity of substitution. The third is motivated by the fact that exchange rates are uncertain. Table 11 shows the detailed description of alternative scenarios.

Table 11. Assumptions under different alternative scenarios
Scenario Assumptions (all changes are compared with the baseline scenario)
Ssa 20% higher of energy efficiency improvement
Ssb 20% lower of energy efficiency improvement
Ssc Elasticity of substitution between labor and capital-energy bundle increase by 20%
Ssd Elasticity of substitution between labor and capital-energy bundle decrease by 20%
Sse Elasticity of substitution between capital and energy increase by 20%
Ssf Elasticity of substitution between capital and energy decrease by 20%
Ssg RMB appreciated by 12%
Ssh RMB appreciated by 8%

It can be seen from Table 12 , lower energy efficiency improvement will require more energy use and hence more CO2 emissions. CO2 mitigation effects and energy saving rates are quite sensitive to the energy efficiency improvement rate. However, the industry is not significantly affected. Moreover, the results suggest that changes in elasticity of substitution between labor and capital-energy bundle have a larger impact than elasticity of substitution between capital and energy. Whereas, the influence of the exchange rate is relatively moderate.

Table 12. Key variables under the baseline and alternative scenarios (%)
Baseline Ssa Ssb Ssc Ssd Sse Ssf Ssg Ssh
CO2 emission change rate compared with the baseline 6.09 –6.69 –4.99 5.89 1.76 –2.05 –0.64 0.59
Energy consumption change rate compared with the baseline 4.75 –5.17 –4.36 5.18 2.17 –2.47 –0.62 0.56
Carbon intensity decreasing rate from 2005 31.79 35.94 27.22 28.38 35.81 32.99 30.39 31.35 32.19
Primary industry proportion 8.03 8.00 8.06 7.98 8.08 8.03 8.03 7.98 8.08
Secondary industry proportion 47.98 48.07 47.89 48.37 47.56 47.91 48.07 47.75 48.20
Tertiary industry proportion 43.99 43.93 44.05 43.65 44.35 44.07 43.90 44.27 43.72

## 7. Discussion and conclusions

This paper provides an analysis of the impacts of CO2 mitigation and economic effects under different carbon tax policies by using a dynamic energy-environment-economy CGE model.

(1) Carbon tax is one of the important policy choices to stimulate the realization of the CO2 intensity target for 2020. Compared with the baseline scenario, carbon tax rate of 30, 60, and 90 RMB per ton CO2 will reduce CO2 emission by 4.52%, 8.59%, and 12.26% in 2020, respectively, under the condition that carbon tax revenues belong to the government. The reduction will lower CO2 emission per unit of GDP by 34.79%, 37.49%, and 39.92% in 2020 with energy efficiency improvement. However, the simulations do not consider the CO2 mitigation from non-fossil energy developments. This paper preliminary estimates the contribution to CO2 abatement from structural energy changes based on the projection of electricity generation by IEA [2009] , along with electricity composition in different pathways by Jiang et al. [2009] . The results show that the proportion of thermal power generation to total power generation drops to 70% and 65% in 2020. This will bring an additional CO2 abatement of 0.56 Gt and 0.85 Gt. Hence, if the energy efficiency improvement and CO2 mitigation from structural energy change can be achieved, the policy of levying a 60 RMB per ton CO2 carbon tax can promote the realization of the 2020 carbon intensity goal.

(2) The negative impacts on GDP from carbon tax are not obvious for the reason that the tax revenue is put into the economic system that can offset GDP loss to some extent. Relative to the GDP of the baseline scenario, the GDP loss rate is 0.11%, 0.25%, and 0.39% in 2020 with the carbon tax rate of 30, 60, and 90 RMB per ton CO2 if the carbon tax revenue is treated as government income.

(3) Imposing carbon tax will have adverse impacts on energy production, energy intensive sectors, and on household income. Levying a carbon tax and reducing the other tax rates of more vulnerable sectors can alleviate the negative impacts on sectors, while CO2 abatement effects are less obvious. Using carbon tax as household income will have greater impacts on the GDP, whereas it can improve the household income level.

(4) Carbon tax will place restrictions on energy demand through an increasing price signal. A carbon tax of 60 RMB per ton CO2 equals to a raise in costs of 109, 168, 180, 176, 182, 186, and 130 RMB per ton of raw coal, coke, crude oil, gasoline, kerosene, diesel oil, and per 1,000 m3 of natural gas, respectively. However, the rigid demand of energy in China will lead to the result that the increasing costs owing to carbon tax are transferred to the energy demand side. This possibly brings uncertainty of CO2 mitigation effects.

The model used in this paper is based on the assumption of perfect competition. It exists deviations to reality, which may bring some bias with the simulations. The parameters selections have significant influence on the simulation results. However, some data are difficult to get or estimate, such as the capital composition coefficient and the elasticity of substitution. Hence, it is necessary to proceed in the calibration and improvement of the model in future work.

## Acknowledgements

The project was supported by National Natural Science Foundation of China (No. 70941034), and “Chinese Environmental Tax” Project of Peking University-Lincoln Institute Center for Urban Development and Land Policy.

## References

1. AGDRCSC (the Academic Group of Development Research Center of State Council, 2005 AGDRCSC (the Academic Group of Development Research Center of State Council; The prospects of China’s economic growth; Decision and Information (in Chinese) (8) (2005), p. 4
2. Bovenberg and de Mooij, 1994 A.L. Bovenberg, R.A. de Mooij; Environmental levies and distortionary taxation; American Economic Review, 84 (1994), pp. 1085–1089
3. Bovenberg and van der Ploeg, 1994 A.L. Bovenberg, F. van der Ploeg; Environmental policy, public finance, and the labour market in a second best world; Journal of Public Economics, 55 (3) (1994), pp. 349–390
4. Cao, 2007 J. Cao; Essays on Environmental Tax Policy Aanalysis; Dynamic Computable General Equilibrium Approaches Applied to China, Harvard University (2007), p. 239
5. Cao, 2009 J. Cao; Walk to a low-carbon development pathway: Carbon tax policy design and CGE model analysis in China; Journal of Financial Research (in Chinese), 12 (354) (2009), pp. 19–29
6. EIA (Energy Information Administration), 2009 EIA (Energy Information Administration); International energy outlook 2009 (2009), p. 274
7. Garbaccio et al., 1999 R.F. Garbaccio, M.S. Ho, D.W. Jorgenson; Controlling carbon emission in China; Environment and Development Economics, 4 (4) (1999), pp. 493–518
8. He, 2001 J. He; A long-term model to forecast population and economy in China; Quantitative and Technical Economics Research (in Chinese) (9) (2001), pp. 40–44
9. He et al., 2002 J. He, K. Shen, S. Xu; A CGE model for carbon tax and CO2 emission reduction analysis  ; Quantitative and Technical Economics Research (in Chinese) (10) (2002), pp. 39–47
10. IEA (International Energy Agency), 2009 IEA (International Energy Agency); World energy outlook 2009 (2009), p. 691
11. IPCC, 2006 IPCC; 2006 IPCC guidelines for national greenhouse gas inventories; ,in: H.S. Eggleston (Ed.), et al. , energy2, , IGES, Japan (2006)
12. Jiang and Hu, 2009 K. Jiang, X. Hu; China’s low carbon development scenario in 2050; ,in: Emissions (Ed.), China Energy and CO2 Emission Report (in Chinese), Study Team of 2050 China Energy and CO2 , Science Press (2009), pp. 753–820
13. Kitous et al., 2010 A. Kitous, P. Criqui, E. Bellevrat, et al.; Transformation patterns of the worldwide energy system-scenarios for the century with the POLES model; Energy Journal, 31(Special Issue on The Economic of Low Stabilization) (2010), pp. 49–82
14. Liang et al., 2007 Q. Liang, Y. Fan, Y. Wei; Carbon taxation policy in China; How to protect energy- and trade-intensive sectors? Journal of Policy Modeling, 29 (2) (2007), pp. 311–333
15. LGSCFNEC (The Leading Group of the State Council on the First National Economic Census, 2004 LGSCFNEC (The Leading Group of the State Council on the First National Economic Census; China Economic Census Yearbook 2004 (in Chinese)  ; China Statistics Press (2004), p. 449
16. Lofgren et al., 2002 H. Lofgren, R.L. Harris, S. Robinson; A standard computable general equilibrium (CGE) model in GAMS; IFPRI (International Food Policy Research Institute) (2002), p. 69
17. NBSC (National Bureau of Statistics of China), 2008 NBSC (National Bureau of Statistics of China); China Statistical Yearbook 2008 (in Chinese)China Statistics Press (2008), p. 1026
18. NBSC (National Bureau of Statistics of China), 2009 NBSC (National Bureau of Statistics of China); 2007 Input-Output Tables of China (in Chinese)China Statistics Press (2009), p. 493
19. NBSC (National Bureau of Statistics of China), and CDNEA (Comprehensive Department of National Energy Administration), 2008 NBSC (National Bureau of Statistics of China), and CDNEA (Comprehensive Department of National Energy Administration); China Energy Statistical Yearbook 2008 (in Chinese)China Statistics Press (2008), p. 284
20. NPDSR (National Population Development Strategy Research), 2007 NPDSR (National Population Development Strategy Research); Research report on national population development strategy; Population Research (in Chinese), 31 (1) (2007), pp. 4–9
21. Paltsev et al., 2005 S. Paltsev, J.M. Reilly, H.D. Jacoby, et al.; The MIT emission prediction and policy analysis (EPPA) model: Version 4. Report No. 125 (2005), p. 72
22. Pan et al., 2001 W. Pan, Z. Li, W. Zhang; Prospects for China economy growth in the first two decades of the 21st century — An analysis based on the models of a supply-oriented economy and a demand-oriented economy; Forecasting (in Chinese), 20 (3) (2001), pp. 1–4
23. Parry, 1995 I.W.H. Parry; Pollution taxes and revenue recycling; Journal of Environmental Economics and Management, 29 (1995), pp. 64–77
24. Peterson et al., 2011 E.B. Peterson, J. Schleich, V. Duscha; Environmental and economic effects of the Copenhagen pledges and more ambitious emission reduction targets; Energy Policy, 39 (6) (2011), pp. 3697–3708
25. Shi and Zhou, 2010 M. Shi, S. Zhou; The impact of low carbon technology development on China’s CO2 emission reduction  ; Management Review (in Chinese), 22 (6) (2010), pp. 48–53
26. Su et al., 2009a M. Su, Z. Fu, W. Xu, et al.; Forecasting and impact assessment of carbon tax of China; Environmental Economy (in Chinese) (9) (2009), pp. 23–27
27. Su et al., 2009b M. Su, Z. Fu, W. Xu, et al.; International experiences and enlightenment of carbon tax; Environmental Economy (in Chinese) (9) (2009), pp. 28–32
28. Wang et al., 2005 C. Wang, J. Chen, J. Zou; Impact assessment of CO2 mitigation on China economy based on a CGE model  ; Journal of Tsinghua University (Science and Technology) (in Chinese), 45 (12) (2005), pp. 1621–1624
29. Wang et al., 2009 J. Wang, G. Yan, K. Jiang, et al.; The study on China’s carbon tax policy to mitigate climate chang; China Environmental Science (in Chinese), 29 (1) (2009), pp. 101–105
30. Wei et al., 2008 Y. Wei, L. Liu, Y. Fan, et al.; China Energy Report: CO2 Emission Research (in Chinese)Economic Science Press (2008), p. 236
31. Wu and Xuan, 2002 Y. Wu, X. Xuan; Economic Theory of Environmental Tax and Analysis of Their Application in China (in Chinese)Economic Science Press (2002), p. 458
32. Zhang, 1998 Z. Zhang; Macroeconomic effects of CO2 emission limits: A computable general equilibrium analysis for China  ; Journal of Policy Modeling, 20 (2) (1998), pp. 213–250

## Notes

. Output in 2020 mentioned here are calculated based on constant price of the base year (2007)

. Output in 2020 mentioned here are calculated based on constant price of the base year (2007)

. Simulated by Qiang Liu from ERI

. Carbon tax are implemented in different time among countries, and here using the average exchange rate in 2007, that is, 1 Euro=10.42 RMB by exchange rate

. 1 RMB=0.133 US\$ by exchange rate in 2007

### Document information

Published on 15/05/17
Submitted on 15/05/17

Licence: Other

### Document Score

0

Views 6
Recommendations 0