## Abstract

The intensity allocation criteria of carbon emissions permits and its influence on China’s regional development are analyzed through the 30-province/autonomous region computable general equilibrium (CGE) model. Simulation results show that: industrial intensity criteria without taking regional economic development into account deepen the unbalance of regional economic development; regional intensity criteria without taking industrial properties into account exert little negative impact on regional harmonious development, but relatively high negative influence on high-carbon emission industries. The two-step allocation scheme that the central government allocates emissions permits to provincial governments based on regional economic development and then provincial governments allocate emissions permits to emission resources or entities based on industrial properties is a feasible and operable choice.

## Keywords

carbon emissions permits ; intensity allocation criteria ; regional balanced development ; computable general equilibrium model

## 1. Introduction

A carbon emissions trading mechanism which stimulates enterprises to reduce carbon emissions through the emissions permits trading price fluctuations can achieve certain emissions reduction targets effectively. In general, a carbon emissions trading mechanism has three steps: firstly setting an emissions cap to make carbon emissions permit a scarce resource; secondly allocating carbon emissions permits to entities or sources of emissions; thirdly entities or sources of emissions transact carbon emissions permits according to their costs and benefits of carbon emissions reduction in the secondary market in which the price of carbon permits is formed.

The initial allocation of carbon emissions permits can cause structural adjustment and income distribution effects. It has a great influence on the political feasibility and economic efficiency of a carbon trading system and is the core issue of designing a carbon emissions trading system. The initial allocation scheme of carbon emissions permits program includes allocation structure, allocation method, allocation criteria, and the use of permit revenue. Allocation criteria which determine the volume of carbon emissions permits that the entity or emissions source can get are the core issue of the carbon emissions permits allocation scheme. Allocation criteria can be divided into absolute allocation criteria and intensity allocation criteria. The volume of carbon emissions permits that the entity or emissions source can get for free is fixed before the running of the trading mechanism. It has nothing to do with their current performance such as grandfathering emissions permits in an absolute allocation criteria scenario, while in an intensity allocation criteria scenario, it is set according to their output or value added such as benchmarking and output based allocation. Intensity allocation criteria are more fair, easily implemented, and can reduce carbon leakage risk and have higher emissions reduction efficiency than absolute allocation criteria [ Demailly and Quirion, 2006 , Bernard et al., 2007 , Fischer and Alan, 2007  and Bushnell and Chen, 2009 ].

The Chinese Government has issued a clear signal to adopt a carbon emissions trading scheme. How to allocate carbon permits will have great impacts on the regional economic development and industrial structure. Only when all these impacts are taken into account, cap and trade policy can be implemented politically and effectively. A two-stage allocation structure is adopted in many occident cap and trade programs such as the EU ETS (European Union’s Emissions Trading System) and RGGI (Regional Greenhouse Gas Initiative). In the first stage, the Committee should allocate carbon emissions permits to local memberships according to the level of regional economic development and CO2 emissions levels. In the second stage, local memberships should allocate carbon emissions permits to industrial sources of emissions or entities according to industrial attributes. The two-stage allocation structure has a great advantage that it can take both industrial properties and regional harmonious development into account. That advantage makes it feasible in China.

Yuan et al. [2011] shows that intensity allocation criteria have many advantages such as lower carbon leakage risk and higher economic efficiency than absolute allocation criteria in China. However, the previous studies [ Fisher , 2003  ; Quirion , 2009 ] mainly did research on how to allocate carbon permits among industries and rarely considered regional allocation and initial allocation on regional economic development and regional equity. China has a vast territory and regional economic disparities. Emissions reduction policies need to take regional harmonious development targets into account. Will it increase the unbalance of regional economic development by setting the intensity allocation criteria only according to industrial properties? What will it be if intensity allocation criteria are set by the level of regional development? Is it possible to use the initial allocation of carbon emissions permits to alleviate regional economic inequity? This paper adopts a multi-regional computable general equilibrium (CGE) model, and simulates the impacts of the intensity allocation criteria on China’s regional economic development to provide policy references for the design of carbon emissions trading mechanisms in China.

## 2. Model

### 2.1. A brief introduction of a 30-province CGE model in China

A multi-regional CGE model is evolved from a single-regional CGE model. Different from a single regional CGE model, a multi-regional CGE model has a module that describes economic linkages and interactions across regions such as interregional commodity flow, labor flow, and capital flows, and so on. It is an effective tool for the analysis of regional allocation of carbon emissions permits and the impacts on regional economic development. The database is the interregional input-output table in 2002 which includes 30 provinces (except Tibet, Hong Kong, Macao, and Taiwan) and 60 industries developed by Research Center on Fictitious Economy & Data Science, Chinese Academy of Sciences. The maximum number of equations that GAMS software can solve is 500,000. Industries are merged in order to make the model sovable. The model contains 30 regions, 19 industries, and more than 444,000 equations.

### 2.2. Carbon emissions permits initial allocation modular

The supply of carbon emissions permits is determined by emissions reduction targets and the emissions cap. Regulated enterprises or departments must offset the same volume of emissions permits that they emit. These are the demand of carbon emissions permits. The price of carbon permits is set when the supply and demand are equal. In the model, the carbon emissions permits are regarded as a production factor which exerts direct influences on the production cost and these influences can be multiplied into the whole economic system through supply-demand, industrial and regional economic relations.

The assumption of the model is perfect information and competition, therefore the trading price of carbon emissions permits is equal to the auction price. Assume that CP is the auction price of carbon emissions permits, and CIN is the national carbon emission intensity target; r represents region and j represents industry, jE represents regulated industry and there are R regions, J industries, and I regulated industries in sum in the model. PPr,j E , OPr,j E , VADr,j E , and CEr,j E represent the producer price of products, the output, the added value, and the real carbon emissions of industry jE in region r . CR represents the carbon emissions permits volume auctioned in the market. CBr,j E represents the volume of carbon emissions permits that is allocated to a unit of value added of industry jE in region r for free. CBr,j E × VADr,j E is the volume of carbon emissions permits that industry jE in region r can get for free. Thus:

 ${\displaystyle {PP}_{r,j_{E}}\times {OP}_{r,j_{E}}=othercost+CP\times \left({CE}_{r,j_{E}}-\right.}$${\displaystyle \left.{CB}_{r,j_{E}}\times {VAD}_{r,j_{E}}\right){\mbox{;}}}$
( 1)

 ${\displaystyle \sum _{r=1}^{R}\sum _{j=1}^{J}{CE}_{r,j}\leq CIN\times \sum _{r=1}^{R}\sum _{j=1}^{J}{VAD}_{r,j}{\mbox{;}}}$
( 2)

 ${\displaystyle CR=\sum _{r=1}^{R}\sum _{j_{E}=1}^{I}{CE}_{r,j_{E}}-\sum _{r=1}^{R}\sum _{j_{E}=1}^{I}{CB}_{r,j_{E}}\times {VAD}_{r,j_{E}}{\mbox{.}}}$
( 3)

Eq. (1) represents the demand of carbon emissions permits and the production function of industry jE in region r. CEr,j E is the volume of carbon emissions permits that industry jE in region r needs to offset for its carbon emissions, and also the real carbon emissions of industry jE in region r . |CEr,j E – CBr,j E × VADr,j E | is the volume of carbon emissions permits that industry jE in region r needs to buy (CEr,j E – CBr,j E × VADr,j E < 0) or sell (CEr,j E – CBr,j E × VADr,j E < 0). Eq. (2) is the cap of carbon emissions trading in order to realize carbon reduction target and represents the supply of carbon emissions permits. Eq. (3) represents the volume of carbon emissions permits auctioned in the market. In the model, CIN and CBr,j E are exogenous while CP and CR are endogenous.

## 3. Scenarios

This paper is to examine the impacts of intensity allocation criteria on our macro-economy and regional development by making static shocks to the baseline scenario without carbon emissions trading mechanism. There are three scenarios.

S0 is the baseline scenario without carbon emissions trading mechanism. Since the Chinese Government has set the year 2005 as the base year of carbon emissions mitigation target, we make the 30-province CGE model dynamic from 2002 to 2005 by the parameter calibration, taking the actual situation of 2002–2005 as a reference for history matching, and then make static policy shock to the base year 2005.

Two scenarios (S1, S2) of carbon emissions trading mechanism are set according to the different intensity allocation criteria and other aspects of the initial allocation plan of carbon emissions permits are the same.

(1) The emissions reduction targets and industries covered by the carbon trading mechanism. Shi et al. [2011] shows that in order to achieve the target of carbon emission intensity decreasing by 40% in 2020 compared with 2005, carbon emission intensity needs to reduce by 10% in the carbon emission mitigation policy scenarios compared with the baseline. Therefore, the intensity target of emissions reduction of carbon emissions trading scenarios is set 10% off compared with the baseline scenario. Considering the technical feasibility and corresponding monitoring costs of emissions measurement, reporting and verification, the carbon emissions trading mechanism should cover industries that emit a lot and can be easily monitored with a soft monitoring cost such as steel, nonferrous metals, petroleum processing, chemical and thermal power generation, and so on.

(2) Allocation method. Carbon emissions permits are allocated by a hybrid way of free distribution and auction. To avoid uncertain impacts on economic development, carbon emissions permits are mainly allocated freely. Because the carbon emissions reduction target in China is intensity-based in which the emissions reduction volume and carbon emissions permit cap varies with economic development, the central government should hold a certain number of carbon emissions permits as a reserve and put them into market by auction according to the extent of realization of the emission-intensity reduction target and economic growth. That means the volume of carbon emissions permits allocated by auction is determined by intensity emissions reduction target and economic growth.

(3) Use of permit auction revenue. Because an emissions trading mechanism usually covers some industries and regions, the auction revenues should be used to reduce the negative impacts on the industries and regions covered by mechanism for the sake of fairness [ Tietenberg , 2003 ]. Many greenhouse gases emissions trading mechanisms allocate most of carbon emissions permits to enterprises for free, which are recessive subsidies to enterprises. Free allocation and excessive subsidies often make enterprises win one-time windfall profits [ Bovenberg and Goulder, 2000 , Catherine and Quirion, 2002  and Burtraw et al., 2005 ]. The cost of carbon emissions permits may be passed through to the consumers via price leverage because of the rigid demands of the products of energy and high carbon emissions. So the auction revenues of carbon emissions permits can be returned to consumers for reducing the negative impacts on them [ Sijm et al., 2006 , Chen et al., 2008  and Fell, 2008 ]. In this paper, the auction revenue is returned to consumers.

(4) Intensity allocation criteria. On the one hand, for reducing the impacts on covering enterprises and improving the political feasibility of mechanisms, the intensity allocation criteria should be set as high as possible. On the other hand, the intensity allocation criteria cannot be too high in order to ensure the scarcity of carbon emissions permits and realize the emissions reduction target. If setting the intensity allocation criteria as that of the base scenario multiplied by 0.9, the model has no feasible solution because the intensity allocation criteria are set too high to ensure the scarcity of carbon emissions reduction and the realization of reduction targets. It is appropriate to set the intensity allocation criteria as that of the base scenario multiplied by 0.8. According to the different objects set by the intensity criteria, two initial allocation scenarios of carbon emissions permits are set. In S1 scenario, the intensity allocation criteria are set according to the industry attributes, the same industry across regions has the same emissions intensity criteria and different industries have different intensity allocation criteria, which are equal to industrial emission intensity of regulated industries in baseline scenario multiplied by 0.8. In S1, the power industry obtains higher intensity allocation criteria for free permits. In S2, the intensity allocation criteria are set according to regional emission intensity, all industries in the same region have the same emission intensity allocation criteria, and different regions have different intensity allocation criteria, which are equal to regional emissions intensity of regulated industries in the baseline scenario multiplied by 0.8. The intensity allocation criteria of the undeveloped regions and resource-intensive regions are relatively higher, while the developed regions are lower.

## 4. Simulation results

### 4.1. Transaction of carbon emissions permits and regional costs

The shadow price of carbon emissions permits is 277.18 RMB per ton CO2 , which is higher than the carbon price of EUA (EU allowance) in S1, and is 135.72 RMB per ton CO2 in S2, which is a little higher than the price of CER (certified emissions reduction) in EU carbon market and lower than the transaction price of EUA. In S1 and S2, the volumes of carbon permits allocated for free are 3.018 billion and 3.012 billion t CO2 respectively, and account for 90.69% and 91.52% of the total volume of emissions permits. The auctioned carbon permit proportion is 9.31% and 8.48% respectively in S1 and S2, which are close to that of the EU ETS in the second trading period (10%).

The minimum trans-regional transaction volume is defined as the sum permit of net sales of the netseller regions. The minimum trans-industrial transaction volume is equivalent to the sum of carbon emissions permits that are net sold out by net-seller industries. The minimum trans-regional and transindustrial transaction volume is the sum of carbon emissions permits net sold out by the net sellers of industries of the regional level. In S1 and S2, the minimum trans-regional transaction volumes are respectively 0.186 billion and 0.028 billion t CO2 , the minimum trans-industry transaction volumes are respectively 0.000 and 0.942 billion t CO2 , and the minimum trans-regional and trans-industry transaction volumes are 0.356 billion and 1.070 billion t CO2 . The minimum trans-industrial transaction volume is lower in S1 because its allocation criteria are set according to industrial differences. The minimum trans-regional transaction volume is lower in S2 because its allocation criteria are set considering regional differences. Carbon intensity differences among regions are greater than those among industries, which leads to a higher volume of the trans-regional and trans-industry transaction of carbon emissions permits in S2.

The transaction amount is equivalent to the transaction volume of carbon emissions permits multiplied by the trading price. In S1 and S2, the minimum trans-industrial and trans-regional transaction amounts are respectively 184.70 billion and 183.17 billion RMB, and account for 1.033% and 1.026% of GDP respectively.

Enterprises should offset equivalent emissions permits that they emit. When a region is a net seller of carbon emissions permits, its carbon emissions cost is negative and it gets some profits by selling carbon emissions permits. In S1, the ratios of carbon emissions permits costs to GDP among regions are of great disparities (Table 1 ). The ratios of carbon emissions permits costs to GDP are higher in resource-intensive provinces and underdeveloped provinces than those in developed provinces. That the ratio is negative means the region is the net seller of carbon emissions permits who can get benefits by selling carbon emissions permits. The number of net seller provinces is great in western area. The ratios of carbon emissions permits costs to GDP among regions are similar in S2.

Table 1. Regional ratio of carbon emissions permits costs to GDP and change in major indicators in China (unit: %)
Region Ratio of carbon emissions permits costs to GDP Change in carbon regional GDP Change in emission intensity Change in carbon emissions
S1 S2 S1 S2 S1 S2 S1 S2
Beijing 0.38 0.19 –0.63 –0.84 5.51 –4.50 4.84 –5.30
Tianjin 0.61 –0.54 –4.10 –2.38 –43.41 –30.29 –45.73 –31.95
Hebei 2.91 0.48 –2.85 –1.05 –14.19 –8.27 –16.63 –9.24
Shanxi 4.97 –0.11 –7.25 0.53 –28.25 –14.08 –33.45 –13.63
Inner Mongolia 3.35 0.49 –8.07 –1.69 –29.95 –10.66 –35.60 –12.17
Liaoning 0.97 0.18 –0.52 –0.63 –11.64 –12.98 –12.10 –13.53
Jilin 1.07 –0.31 –1.17 –0.91 –20.03 –21.07 –20.96 –21.79
Heilongjiang 1.25 0.34 –0.96 –0.53 –5.73 –6.49 –6.63 –6.98
Shanghai –0.63 0.21 1.14 0.07 5.22 –6.36 6.43 –6.29
Jiangsu –0.08 0.22 0.55 –0.47 –0.25 –10.28 0.30 –10.70
Zhejiang –0.42 0.26 1.24 –0.17 3.74 –5.03 5.03 –5.20
Anhui 1.67 0.41 –0.69 –0.33 –14.16 –7.27 –14.76 –7.58
Fujian –0.09 0.21 0.76 0.03 5.77 –3.46 6.57 –3.42
Jiangxi 0.10 0.38 0.45 –0.56 0.11 –6.60 0.55 –7.13
Shandong 1.12 0.01 –1.48 –0.61 –19.44 –14.95 –20.63 –15.47
Henan 0.58 –0.08 –0.15 –0.38 –20.19 –17.73 –20.30 –18.05
Hubei –0.74 0.39 1.88 0.13 11.30 –2.07 13.39 –1.94
Hunan 1.53 0.40 –0.51 –0.43 –6.73 –5.13 –7.21 –5.54
Guangdong –0.91 0.21 1.78 0.30 11.90 –6.11 13.89 –5.83
Guangxi 0.40 0.28 1.10 0.06 –1.41 –4.86 –0.32 –4.80
Hainan 0.81 0.12 –0.60 –0.48 6.03 –7.15 –6.59 –7.60
Chongqing –0.36 0.50 1.15 –0.08 12.33 –0.48 13.62 –0.56
Sichuan –0.61 0.32 1.64 –0.33 8.49 –2.33 10.27 –2.65
Guizhou 4.20 0.67 –7.83 –1.47 –23.10 –9.56 –29.12 –10.89
Yunnan 2.67 0.66 –1.06 –0.29 –10.03 –4.23 –10.99 –4.51
Shaanxi 1.19 0.03 –1.36 –0.50 –12.51 –12.29 –13.70 –12.73
Gansu 0.74 0.60 0.60 –1.71 –3.03 –8.89 –2.44 –10.45
Qinghai –3.95 0.14 5.14 0.26 52.80 1.69 60.64 1.96
Ningxia 5.74 –0.21 –12.13 –2.15 –38.27 –16.92 –45.76 –18.71
Xinjiang 2.22 0.05 –2.95 –0.66 –24.25 –14.02 –26.48 –14.58
Total 0.48 0.23 –0.25 –0.39 –10.00 –10.00 –10.23 –10.35

Note: Tibet, Hong Kong, Macao, and Taiwan are not included in the model because of great difficulties in data availability

### 4.2. Impacts on economic welfare and regional disparity

The loss rates of national GDP in S1 and S2 are 0.25% and 0.39% respectively, both of which are controlled within 0.50%.

The influences on regional economic development are of great difference between S1 and S2. The coefficients of variation in regional GDP per capita in S0, S1 and S2 are respectively 0.1243, 0.1280 and 0.1248, and the imbalance of regional development in S1 is intensified. Compared with S1, the influences on GDP among regions in S2 are relatively even (Table 1 ). In S1, the loss rates of GDP in resource-intensive provinces such as Ningxia, Inner Mongolia, Guizhou, and Shanxi are more than 7%; the loss rates of GDP in Tianjin, Xinjiang, Hebei, Shaanxi, Jilin, Yunnan are between 1% and 3%. Regional GDP increases in some developed provinces such as Shanghai, Jiangsu, Zhejiang, Fujian, Guangdong. In S2, the extent of negative influences on resource-intensive provinces and the extent of positive influences on some developed provinces are both reduced.

The differences of impacts on regional GDP among regions result from different regional intensity allocation criteria which have a great influence on the regional carbon emissions permits costs in S1 and S2. The change rates of regional GDP are significantly in negative correlation with the ratios of carbon emissions permits costs to GDP (Fig. 1 ). This means the higher the ratio of carbon emissions permits costs to regional GDP, the greater negative impacts on regional GDP.

 Figure 1. Relationship between the change of GDP and the ratios of carbon emissions permit cost to GDP

### 4.3. Impacts on regional CO2 emissions

The CO2 emission intensity in China is reduced by 10% in both scenarios S1 and S2 compared with the baseline scenario (Table 1 ).

Relatively speaking, there are greater disparities of CO2 emissions reduction among regions in scenario S1 compared with S2. The emissions reduction rates of 9 provinces such as Tianjin, Shanxi, Inner Mongolia, Jilin, Shandong, Henan, Guizhou, Ningxia, and Xinjiang are more than 20%. However, the CO2 emissions in Beijing, Shanghai, Jiangsu, Zhejiang, Fujian, Jiangxi, Hubei, Guangdong, Chongqing, Sichuan, and Qinghai increase, and the growth rate of carbon emissions in Guangzhou reaches 13.89%. Among the provinces except Jiangsu in which the carbon emissions increase, the intensity of carbon emissions also increases. The emissions reduction pressure of resource-intensive provinces in the midwest, the carbon emissions and emission intensity in developed regions are lower in scenario S2 than those in scenario S1.

### 4.4. Impacts on industrial structure

The regulated industries that are all emissionintensive must pay a permit cost for their carbon emissions in the cap and trade mechanism. The negative influences are multiplied to energy industry through industrial relation. Therefore, the output loss rates of energy and regulated industries are high while those of service and agriculture and the other industries are relatively small (Table 2 ).

Table 2. Change of output in different scenarios compared with the baseline in China (unit: %)
Region S1 S2
I1 I2 I3 I4 I5 I6 I1 I2 I3 I4 I5 I6
Beijing –2.6 –3.0 –1.1 –2.9 –3.1 –1.7 –2.3 –6.6 –5.2 –2.4 –3.3 –1.5
Tianjin –3.2 –5.3 –13.2 –4.8 –2.1 –3.4 –2.1 –3.9 –6.1 –3.4 –0.6 –2.5
Hebei –2.6 –6.0 –10.3 –3.6 –4.1 –1.6 –1.3 –7.3 –3.1 –1.6 –1.6 –0.7
Shanxi –3.1 –6.9 –20.0 –9.7 –6.8 –3.8 0.4 –2.5 5.1 0.2 0.2 –1.5
Inner Mongolia –3.0 –24.9 –30.8 –10.1 –6.1 –4.2 –0.7 –9.9 –3.4 –3.8 0.9 –1.5
Liaoning –0.7 –2.3 –1.5 –1.3 –0.6 –0.9 –0.6 –2.3 –1.8 –1.4 –0.9 –0.8
Jilin –0.8 –7.8 –5.4 –1.8 –0.2 –0.5 –0.1 –7.9 –4.0 –1.5 –0.1 –0.4
Heilongjiang –0.4 –1.3 –4.1 –1.9 –1.2 –1.0 0.1 –1.4 0.6 –1.6 –0.5 –0.9
Shanghai 0.1 1.7 5.7 0.6 0.7 0.3 –0.8 –1.7 –1.5 –0.7 –0.1 –0.3
Jiangsu 0.2 –0.1 2.1 0.2 0.1 0.2 –0.2 –5.8 –3.0 –0.7 –1.1 –1.0
Zhejiang 1.0 2.6 5.7 1.6 0.8 0.5 –0.5 –4.1 –1.5 –0.5 –0.3 –0.4
Anhui –0.2 –1.3 –8.7 –0.6 –1.5 –0.2 –0.1 –3.8 –3.2 –1.0 –0.7 –0.3
Fujian 1.3 0.0 3.1 0.6 0.5 0.7 0.3 –0.1 –1.7 –0.3 –0.2 –0.2
Jiangxi 1.1 3.2 –2.6 0.2 –0.4 0.9 0.3 –6.6 –2.7 –1.6 –0.8 –0.4
Shandong 0.0 –3.4 –7.1 –2.5 –1.5 –1.2 0.3 –3.0 –2.5 –1.0 –0.8 –1.0
Henan 0.6 –2.9 –2.4 –1.2 –1.2 –0.4 –0.2 –4.5 –2.4 –1.3 –1.4 –0.6
Hubei 0.3 15.2 6.3 1.9 1.3 1.0 0.1 0.9 –0.5 –0.2 –0.4 –0.1
Hunan 0.0 –1.2 –6.0 –0.8 –1.6 –0.3 –0.1 –2.3 –4.1 –1.1 –1.2 –0.4
Guangdong 1.6 0.5 8.1 2.4 1.9 1.0 –0.3 –0.4 –0.3 0.1 –0.1 –0.5
Guangxi 1.3 7.6 3.4 0.5 0.3 0.6 0.8 –8.0 –0.7 –0.4 –1.0 –0.5
Hainan 0.0 –14.4 –8.6 –0.2 –2.9 –0.5 –0.2 –8.8 –3.7 –1.3 –3.0 –0.8
Chongqing 0.1 13.3 5.0 0.9 –0.6 0.6 –0.3 –0.5 –1.2 –0.8 –0.6 –0.3
Sichuan 0.9 4.9 7.0 1.3 1.2 1.0 0.0 –3.0 –2.0 –0.9 –2.0 –0.4
Guizhou –1.9 –30.4 –27.4 –7.5 –7.6 –1.5 0.7 –11.7 –5.4 –3.1 –3.0 –0.5
Yunnan –0.1 –16.3 –10.7 –1.7 –3.0 –0.4 –0.1 –6.0 –3.6 –1.0 –1.3 –0.4
Shaanxi 0.5 –1.8 –5.8 –1.9 –0.8 –0.6 0.7 –1.1 –0.3 –1.2 –1.3 –0.9
Gansu 0.8 19.8 1.0 0.5 0.7 0.7 0.1 –2.3 –5.8 –3.5 –2.0 –0.7
Qinghai 0.9 5.0 43.0 5.7 6.8 6.0 –0.1 4.2 0.4 –1.7 –0.6 –0.2
Ningxia –1.9 –20.1 –32.6 –11.5 –13.7 –2.8 0.0 –7.2 –2.2 –3.5 –5.3 –1.0
Xinjiang –2.9 –1.6 –14.9 –3.8 –4.8 –2.0 –1.0 –0.5 –2.4 –2.6 –0.5 –0.9
Total –0.1 –3.2 –1.3 –0.2 –0.8 –0.3 –0.2 –3.1 –2.1 –0.8 –0.8 –0.7

Notes: I1–I6 represents agriculture, energy, regulated industries, other industries, transportation and storage, and other service industries respectively. Tibet, Hong Kong, Macao and Taiwan are not included in the model because of great difficulties in data availability

In S1, the output loss rates of energy and regulated industries are of great disparities among regions. The output loss rate is higher in the midwest area than that of eastern developed area. In view of regional level, the output loss rates of energy in the provinces such as Inner Mongolia, Guizhou, Ningxia, Yunnan, Hainan, Jilin, and Shanxi, and those of regulated industries in the provinces such as Shanxi, Inner Mongolia, Guizhou, Ningxia, Tianjin, Hebei, Yunnan, and Xinjiang are very high. But the outputs of energy and regulated industries in Shanghai, Zhejiang, Fujian, Hubei, Guangdong, Guangxi, Chongqing, Sichuan, Gansu, and Qinghai increase. The output loss rates of energy and regulated industries among regions in S2 are relatively even compared with S1.

Intensity allocation criteria have considered the industrial properties and the output loss rates are relatively even among regulated industries in S1. The outputs of all regulated industries are reduced: thermal power by 5.63%, petroleum processing industry by 2.13%, and other regulated industries by less than 0.50% compared with the baseline scenario. The output loss rates are of great disparities among regulated industries in S2. The output of thermal power declines by 15.19%; non-metallic mineral products, metal smelting by 0.42% and rolling processing industry by 1.34%, while the output of petroleum processing industry and chemical industry increase by 2.23% and 0.10% respectively compared with the baseline scenario in S2. Intensity allocation criteria are set according to regional development without considering the industrial attributes in S2. That means the relatively more-emission-intensive regulated industries get less carbon permits for free and have to buy more carbon permits from less-emission-intensive regulated industries. As a result, the relatively more-emissionintensive regulated industries’ carbon permits cost increases and output decreases while the relatively lessemission-intensive regulated industries’ carbon permits revenue increases and output increases.

## 5. Conclusions and policy implications

This paper simulates the impacts of intensity allocation criteria of carbon emissions permits on regional economic development. The results indicate: 1) Intensity allocation criteria based on industrial properties and that based on regional economic development have distinctive impacts on regional economic development in China. Industrial intensity allocation criteria deepen the unbalance of regional economic development, while regional intensity allocation criteria exert little negative impact on regional harmonious development. 2) The distinctions of the impacts of the two kinds of allocation criteria can be attributed to the following two reasons: one is the higher trading price for the carbon emissions permits which results in the high expenditure that the midwest regions spend on carbon emissions permits; the other is the higher emissions burden of the resource-intensive provinces (which are mainly located in midwest) in the industrial intensity allocation criteria scenario than in the regional intensity allocation criteria. 3) Under the regional allocation criteria, the emission-intensive industries, especially the thermal power industry, are impacted significantly negatively. Therefore, the tolerance capacity of the emission-intensive industries should be appropriately considered when the local government allocates carbon permits to industries.

As a country vast in territory but uneven in regional economic development, China takes regional harmonious development as the essential target of the regional policies. Therefore, the development of green economy as well as the low-carbon development should be consistent with regional harmonious development. In the processing carbon emissions trading scheme, the intensity allocation criteria based on regional economic levels are conducive to the win-win between low-carbon development and regional harmonious development. The allocation process should be carried out in two steps: the carbon emissions permits should be firstly allocated to each province by the central government according to regional economic development and then allocated to specified emissions entities according to industrial properties by local governments. The two-step allocation structure is practical and feasible since it balances the emissions burden among regions on the one hand, which helps to remit regional resistance, and evens the impacts on each province on the other hand, which is good for regional harmonious development.

The design of the initial allocation scheme of the carbon emissions trading requires further studies, such as the link between realistic factors and model operations like the link between the two stages and the policy game between the central government and regional governments. Besides, the issues like the longterm impacts of carbon emissions permits allocation on economic development and the optimization on the settings of the emission mitigation path should also be incorporated into consideration.

## Acknowledgements

The project was supported by National Natural Science Foundation of China (No. 71173212, 41101556, and 71203215) and the President Fund of GUCAS (No. Y1510RY00).

## References

1. Bernard et al., 2007 A. Bernard, C. Fischer, A. Fox; Is there a rationale for output-based rebating of environmental levies?; Resource and Energy Economics, 29 (2) (2007), pp. 83–101
2. Bovenberg and Goulder, 2000 A.L. Bovenberg, L.H. Goulder; Neutralizing the adverse industry impacts of CO2 abatement policies; accessed What does it cost? NBER Working Papers, No. 7654 (2000) [2011-04-02] http://www.nber.org/papers/w7654
3. Burtraw et al., 2005 D. Burtraw, K. Palmer, D. Kahn; Allocation of CO2 emissions allowances in the regional greenhouse gas cap-and-trade program; accessed Discussion Paper dp-05-25 (2005) [2010-10-15] http://www.rff.org/RFF/Documents/RFF-DP-05-25.pdf
4. Bushnell and Chen, 2009 J. Bushnell, Y. Chen; Regulation, allocation, and leakage in cap-and-trade markets for CO2; accessed NBER working paper, 15495 (2009) [2010-10-15] http://www.nber.org/papers/w15495.pdf?new_win_dow=1
5. Catherine and Quirion, 2002 B. Catherine, P. Quirion; Implementing greenhouse gas trading in Europe: Lessons from economic literature and international experiences; Ecological Economics, 43 (2) (2002), pp. 213–230
6. Chen et al., 2008 Y. Chen, J. Sijm, B. Hobbs, et al.; Implication of CO2 emissions trading for short-run electricity market outcomes in northwest Europe  ; Journal of Regulatory Economics, 34 (2008), pp. 251–281
7. Demailly and Quirion, 2006 D. Demailly, P. Quirion; CO2 abatement, competitiveness and leakage in the European cement industry under the EU ETS: Grandfathering versus output-based allocation  ; Climate Policy, 6 (2006)
8. Fell, 2008 G.H. Fell; EU-ETS and Nordic electricity: A CVAR approach; accessed RFF Discussion Paper No. 08-31 (2008) [2010-11-12] http://www.rff.org/RFF/Documents/RFF-DP-08-31.pdf
9. Fischer, 2003 C. Fischer; Combining rate-based and cap-and-trade emissions policies; Climate Policy, 3 (S2) (2003), pp. S89–S103
10. Fischer and Alan, 2007 C. Fischer, K. Alan; Output-based allocation of emissions permits for mitigating tax and trade interactions; Land Economics, 83 (4) (2007), pp. 575–599
11. Quirion, 2009 P. Quirion; Historic versus output-based allocation of GHG tradable allowances: A comparison; Climate Policy, 9 (6) (2009), pp. 575–592
12. Shi et al., 2011 M.-J. Shi, Y.-N. Yuan, N. Li, et al.; Policy choices of energy saving and emission reduction facing climate change; Y. Fan (Ed.), Research on Costs , Roads and Policies of Greenhouse Gas Mitigation (in Chinese), Science Press (2011), pp. 152–192
13. Sijm et al., 2006 J. Sijm, K. Neuhoff, Y. Chen; CO2 cost pass-through and windfall profits in the power sector  ; Climate Policy, 6 (2006), pp. 49–72
14. Tietenberg, 2003 T. Tietenberg; The tradable-permits approach to protecting the commons: Lessons for climate change; Oxford Review of Economic Policy, 19 (3) (2003), pp. 400–419
15. Yuan et al., 2012 Y.-N. Yuan, M.-J. Shi, N. Li, et al.; Initial allocation of carbon permits and regional economic development; M.-J. Shi (Ed.), et al. , Policy Choice of Low Carbon Development and Its Regional Response (in Chinese), Science Press (2012), pp. 155–174

### Document information

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

Licence: Other

### Document Score

0

Views 2
Recommendations 0