Integrated impact of the carbon quota constraints on enterprises within supply chain: Direct cost and indirect cost

Renewable and Sustainable Energy Reviews 92 (2018) 774–783

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews journal homepage: www.elsevier.com/locate/rser

Integrated impact of the carbon quota constraints on enterprises within supply chain: Direct cost and indirect cost

T

⁎

Chen Wanga, Zhaohua Wangb,c,d, , Ruo-Yu Keb,c,d, Jiancai Wangb,c a

Donlinks School of Economics and Management, University of Science and Technology Beijing, Beijing 100083, China School of Management and Economics, Beijing Institute of Technology, Beijing 100081, China c Center for Energy & Environmental Policy Research, Beijing Institute of Technology, Beijing 100081, China d Collaborative Innovation Centre of Electric Vehicles in Beijing, Beijing 100081, China b

A R T I C LE I N FO

A B S T R A C T

Keywords: Carbon quota constrains Competitiveness Game theory Supply partnership

Emission Trading Scheme (ETS) has become one of the most popular ways to meet CO2 reduction targets owing to its flexibility and cost effectiveness. This paper applies game theory to analyze the impacts of direct and indirect costs derived from ETS on enterprise's competitiveness and supply partnership. By emphasizing the role of carbon intensity, the results illustrate that an enterprise with lower carbon intensity will have stronger capability to ease the pressures of both direct and indirect costs. From the perspective of direct cost, when reducing carbon intensity to a certain extent, the enterprise will take the advantage of carbon competitiveness to further expand the market share. The enterprise with low carbon intensity can even increase its product output amount instead of damaging it. From the perspective of indirect cost, a downstream manufacturer with lower carbon intensity is more likely to obtain the favor of suppliers. If choosing a downstream partner with high carbon intensity, the supplier may need to reduce the price of its product to ensure the product's demand and profits. Therefore, the implementation of ETS will drive suppliers to choose low-carbon partners, resulting in low-carbon supply partnerships replacing the original one. Based on a numerical example combing with the investigation in Hubei ETS pilot, this research argues for the enlightenment, and implications, of carbon quota constraints as a part of China's emission reduction policies.

1. Introduction Climate change, in the most recent five decades, has been mainly caused by greenhouse gas emissions from anthropogenic activities, according to the fourth report of the International Panel on Climate Change [9]. The greenhouse effect has affected the natural, social, and ecological environment on which human rely, with both its speed and extent already exceeding popular expectation. In response to climate change, various countries have explored many methods, among which carbon emission trading scheme (ETS) has become one of the most popular ways to meet CO2 reduction targets owing to the flexibility and cost effectiveness [11,17,33]. As a developing country with rapid economic growth and the largest CO2 emissions in the world, China's reactions to environmental issues have drawn considerable attention. At the Paris UN Climate Conference, China clearly proposed that it would reduce its carbon intensity by 60–65% by the end of 2030, compared with 2005. To achieve its CO2 emission reductions targets and to identify ways to build a unified carbon emissions trading scheme—by drawing on the experience of developed countries—China has gradually

⁎

established seven pilots since 2012 specifically for Beijing, Shanghai, Tianjin, Chongqing, Guangdong, Hebei, and Shenzhen. In this way, the government has implemented initiatives to put emissions trading plans into effect. Based on practice from these pilot programs, China has also promised to steadily establish a national carbon market. Now at a learning-by-doing stage, there is keen interest from both policy makers and business managers in China to understand the impacts of ETS on an enterprise's production operational strategy. Under ETS, a CO2 permit becomes a kind of commodity that can be exchanged on the carbon market. In an allowance-based ETS, participants face a certain free carbon quota. No extra permit is needed until emissions exceed the free quota. To cover actual emissions, enterprises will start to purchase carbon permits and carbon exchange will occur. Along with carbon price fluctuations, the cost structure will change such that operational decisions will be adjusted correspondingly. Regarding costs, ETS exerts pressure on enterprises by increasing both direct and indirect costs. So-called direct cost arises from the purchase of permits. Moreover, when an upstream supplier associated with a product has a carbon quota constraint, indirect cost will arise.

Corresponding author at: School of Management and Economics, Beijing Institute of Technology, Beijing 100081, China. E-mail address: [email protected] (Z. Wang).

https://doi.org/10.1016/j.rser.2018.04.104 Received 10 January 2017; Received in revised form 5 April 2018; Accepted 19 April 2018 1364-0321/ © 2018 Elsevier Ltd. All rights reserved.

Renewable and Sustainable Energy Reviews 92 (2018) 774–783

C. Wang et al.

manufacturing industries and found that the EU-ETS exerted different influences on different industrial sectors; Kara et al. [10] investigated the impacts of the EU-ETS implementation on energy intensive industries in Finland; Chen et al. [4] studied the implications of CO2 emissions trading for short-term electricity market outcomes in northwestern Europe; Tsai and Yen [27] analyzed the influence of ETS on power units’ operation and dispatch; and Park et al. [16] aimed to assist private companies in establishing proper investment strategies for CDM projects under uncertain energy policies. Finally, Wang et al. [28] examined the effect of ETS on firms’ production planning, such as optimal production quantity, optimal allowance selling price, and maximum profit. Among the studies on competitiveness changes, some scholars have considered the indirect cost caused by ETS. Based on cost structure, CO2 emissions, electricity consumption, and allocated allowances, Tomas et al. [26] used survey data from four representative businesses in the Portuguese chemical industry to analyze the effects of ETS. In this paper, not only the direct cost derived from purchasing quotas but also the indirect production costs derived from the power sector were considered in constructing the cost structure. Chan et al. [3] used panel data from 5873 enterprises in ten countries during the period from 2001 to 2009 to analyze the changes in unit material costs, employment, and revenue before and after the implementation of carbon trading regimes. The analysis results suggested that rising material costs could reflect the costs to fulfill obligations under the ETS, that is, the indirect cost. The determinants of firm's competitiveness have been discussed in the research of Meleo [14], who considered the possibility of passing through the environmental costs onto the final price as an important part of the determinant framework. It is of great significance to study the impact of ETS on the enterprise's competitiveness. On one hand, such an examination can help enterprises who have carbon quota constraints to achieve reasonable production and a mitigation plan, and thus gain maximum benefit therefrom [12]; On the other hand, such an examination can provide a reference for policymakers to design more reasonable and effective carbon trading policies.

The upstream supplier may then transfer part of the cost increment caused by the carbon quota constraints to the price of its product, which will indirectly raise the raw material costs of the downstream manufacturer. If both upstream and downstream enterprises are incorporated in the ETS, the situation becomes more complicated, where the downstream manufacturer will face dual cost pressures as a result of increases in carbon costs and raw material prices. That, ETS affects enterprises from different levels. Owing to the direct cost, carbon intensity becomes a new competitiveness indicator that influences other competitiveness performance indicators, such as production output, market share, and profit. An enterprise with lower carbon intensity then has stronger carbon competitiveness, resulting in a greater advantage in industrial competition. The indirect cost affect the supply partnership between upstream and downstream enterprises by improving the raw material purchasing costs. Currently, the literature has focused on the effects of ETS on regional economies or industries [15,18]; however, few studies have analyzed the impacts of direct and indirect costs caused by ETS on industrial competition and supply cooperation relationships. More important, from the perspective of practical implementation, many enterprises in China's ETS pilots have not realized the impacts of ETS. Moreover, they neglect the importance of carbon management, as they have not realized the government's seriousness in building a carbon market. At the budding stage of ETS, it is important to discuss the influence from both industrial competition and supply partnership perspectives. In this context, this research aims to address the following three questions: (1) How does the direct cost derived from ETS affect an enterprise's competitiveness? (2) Regarding indirect cost, what are the potential effects of ETS on the supply partnership between upstream and downstream enterprises? (3) How can enterprises manage such with the effects? To address these three issues, this research first uses a duopoly model to analyze the impact of ETS on enterprises within the same industry. In this way, the production output change caused by ETS implementation is mainly discussed. Then, a Stackelberg model is applied to analyze the impact of the indirect cost caused by carbon quota constraints, including the pricing strategy of upstream suppliers and production strategy of downstream manufacturers. The answers to these issues can make up for the deficiency of academic research on the impacts of ETS. On the other hand, we hope that this paper can provide some enlightenment for China's enterprises with respect to carbon management.

2.2. The operational decisions regarding the supply chain with carbon quota constraints In recent years, research into the operational decisions regarding the supply chain under the background of carbon constraints has gradually received more attention. On the one and, pressure from carbon constraints motivates businesses to reduce CO2 emissions across the entire supply chain; on the other hand, consumer concern regarding environmental issues and consumers’ desire for environmentally friendly products make enterprises redesign their supply chain within such carbon constraints. Some scholars consider the supply chain to be a whole in studying the optimal strategy [20,24]. For example, Abdallah et al. [1] developed a mixed integer program that minimized emissions throughout the supply chain by considering green procurement with respect to the carbon-sensitive supply chain. Chaabane et al. [2] designed sustainable supply chains under the emission trading scheme, which suggests that the legislation must be strengthened and harmonized at a global level to develop a meaningful environmental strategy. In addition to the research on the optimal operation decision of the whole supply chain under ETS, the interconnected, restrictive relationship between different actors in the supply chain has also been the focus of scholars. By providing a mathematical background for modelling a rational system and for generating solutions in competitive or conflicting situations, game theory method has been widely used to verify the relationship between different actors in the supply chain. Sheu [22] studied the problem of negotiations between producers and reverse-logistics suppliers in cooperative agreements under government intervention. Based on evolutionary game theory, Sikhar et al. [23]

2. Literature review As ETS has received growing attention, many studies have investigated operational decisions to reduce carbon emissions and to maximize benefits under carbon quota constraints. The research related to this paper can be classified into two categories: the impacts of ETS on enterprises’ competitiveness and the impacts of ETS on the operational decisions regarding the supply chain. 2.1. The impact of ETS on enterprises’ competitiveness The ETS has been complained by some energy-intensive industries since firms believe that carbon quota constraints would reduce their competitiveness. In fact, before the implementing of the carbon emissions trading regime, many scholars have analyzed its potential impacts on business competitiveness. At present, the measurement of competitiveness indicators is not entirely consistent in the literature, as studies have examined costs, profits, output, market share, employment, and so forth. Smale et al. [25] evaluated the potential impact of carbon trading on profit, cost, and market share by using a Cournot model; Lund [13] assessed the EU-ETS impact on the cost of energy intensive 775

Renewable and Sustainable Energy Reviews 92 (2018) 774–783

C. Wang et al.

direct cost increment from purchasing CO2 credits or spontaneous-mitigation-induced costs, as enterprises have to meet carbon quota limits; and the indirect cost increment from purchasing materials with a higher price from a supplier that may pass part of its cost increment caused by carbon quota constraints through to their products. Therefore, the costs will increase in the following two ways:

tried to establish coordination between producers and retailers for the purpose of achieving both environmental and commercial benefits. Using the Stackelberg model, Ren et al. [19] investigated the allocation problem of carbon emission abatement targets in a decentralized maketo-order supply chain comprising a manufacturer and a retailer. Huang et al. [7] formulated a three-stage game mode for a multi-level green supply chain with multiple suppliers, with one manufacturer and multiple retailers, in order to investigate the effects of ETS on the product line design, supplier selection, transportation mode selection, and pricing strategies. In addition, via a Stackelberg model where the supplier is the leader, Xu et al. [30] analyzed the coordination mechanism between the supplier and the manufacturer under ETS. Further, they considered the coordination of a dual-channel supply chain including the supplier and retailer under ETS [31]. Zu et al. [34] used a Stackelberg differential game to consider a two-echelon supply chain consisting of one manufacturer and one supplier that aims to increase sustainable profits. All of the above studies have focused on the cooperation mechanism between different parties in the supply chain with carbon quota constraints, for the purpose of maximizing benefits and reducing carbon emissions. The competition within the supply chain under ETS is another interesting issue in the field of operational decision of the supply chain, which is nonetheless of less concern. Some enlightenments are provided by the research of Yang et al. [32], which considered two competitive supply chains consisting of one manufacturer and one retailer under ETS.

(1) Carbon price pe; with a higher carbon price, ETS participants pay more for carbon permits, leading to a higher carbon cost; (2) The price increment of the raw material: Upstream suppliers have pricing power over their own product, i.e., the raw material sold to downstream manufacturers. Downstream manufacturers will decide their optimal output while taking the material price into consideration, as they aim to maximize their profit under prevailing costs and raw material price conditions. With carbon quota constraints, upstream suppliers may pass part of their carbon cost through to the product, resulting in an increasing indirect cost for downstream manufacturers. The direct and indirect costs affect the industrial competition and supply partnership, respectively, which are discussed in this section. We use a duopoly model to quantitatively explore the changes in competitiveness before and after the implementing of the emission trading scheme, and we use a Stackelberg model to solve the issue regarding how the indirect cost affects the supply partnership. 3.1. The impact of direct cost derived from ETS on enterprises’ competitiveness

2.3. Summary of the literature review In reviewing the related literature, we find that although the impact of ETS on enterprise's competitiveness has attracted much attention among scholars, indirect cost has been rarely considered. In addition, few studies have discussed whether the implementation of ETS will lead to changes in the cooperative relationship among parties of the supply chain. Above all, focusing on the downstream manufacturer in the supply chain, this work aims to analyze the impacts of direct and indirect costs derived from ETS and examines strategies to cope with such costs. We first discuss the impact of direct cost on enterprise's competitiveness, which is mainly reflected in optimal output, and we then explore the impact of indirect cost on the cooperation relationship between suppliers and manufacturers to determine whether the implementation of ETS drives monopolistic suppliers to choose a more low-carbon partner, which may result in a more low-carbon supply relationship. Our key contribution lies in the following two aspects. First, our paper contributes to the literature on the impact of ETS by emphasizing the role of enterprises’ carbon intensity. The introduction of ETS makes the carbon intensity a new competitiveness indictor—that is, carbon competitiveness. With different carbon intensities, enterprises have different sensitivities with respect to carbon price pressure. Thus, the carbon competitiveness affects enterprises’ production strategy and changes their competitive position among industrial competitors. Considering the influence of different carbon intensities between enterprises, we highlight carbon intensity as an important independent variable in this paper. Second, this paper aims to analyze the impact of indirect cost derived from ETS on the supply partnership between suppliers and manufacturers, which has been ignored in the aforementioned literature. Our paper aims to provide insights for enterprises, especially the manufacturer in the supply chain, to manage and optimize their operational decisions.

Following the logic in Sartzetakis [21], the carbon market is regarded as separate from the product market. The product market is an oligopoly, while the carbon market is perfectly competitive. As carbon intensity is a new competitiveness indicator that additionally affects other operational decisions of the firm, we take differences in carbon intensity between two enterprises into consideration in the game model. The variables and parameters used in the game model are explained in Table 1. 3.1.1. Model hypothesis For clearly exploring the impact of direct cost derived from ETS, we use a duopoly model to quantitatively explore enterprise's competitiveness changes before and after the implementing of ETS. The duopoly enterprise fi (i = 1, 2) produces under the carbon emissions cap, with the original allocated free quotas efi. In response to the quota constraints, enterprise fi will first reduce emissions sfi by its own effort. As the total abatement increases, however, the marginal abatement cost will rise accordingly. It is not until the marginal abatement cost is equal to the carbon price that enterprise fi change from reducing emissions by itself to purchasing carbon credits to meet its need. We make several Table 1 Nomenclature in the game model for the impact of direct cost. Symbol

Definition

pe efi sfi

Carbon price Free carbon quota amount allocated to enterprise fi CO2 reduction amount by the independent mitigation behavior of enterprise fi Product output of enterprise fi Marginal production cost coefficient of enterprise fi Carbon intensity of enterprise fi Profit of enterprise fi Proportion of free quota accounted for actual emissions for enterprise fi Marginal abatement coefficient for enterprise fi Production cost of enterprise fi Total abatement cost of enterprise fi Carbon emissions of enterprise fi

qfi mf ρfi πfi θfi γfi Cfi TACfi Efi

3. The impacts of ETS on enterprises When both the upstream supplier and downstream manufacturer of a supply chain are subject to carbon constraints, the potential cost increment for the manufacturer will probably arise from two sources: the 776

Renewable and Sustainable Energy Reviews 92 (2018) 774–783

C. Wang et al.

3.1.2. Solutions of the game model for the impact of direct cost Enterprises operate under the environment of carbon emission constraints, and the profit of enterprise fi consist of two parts, where the positive term comes from the sales of products and carbon quotas, while the negative term includes production cost, abatement cost and carbon quotas purchasing. Thus, the profit of enterprise fi is:

assumptions for the game model, as follows: (1) Perfect rationality: Duopoly enterprises are assumed to be perfectly rational, i.e., they aim to maximize their profits. (2) Enterprises are of high emissions with an oligarchic nature, such as high-intensive industries in China. qfi is the product output of enterprise fi. The two enterprises produce and sell the same products in the market. In addition, they face a linear inverse demand function [29], as shown in Eq. (1).

pf (qf 1, qf 2) = a − (qf 1 + qf 2)

πfi = pf (qf 1, qf 2) qfi − Cfi − TACfi − pe ΔE

Enterprise fi chooses the optimal output qfi and abatement amount sfj* to maximize its profits, as shown in Eq. (7).

(1)

MAX qfi, sfi π fi

Because the main focus of this paper is the impact of ETS on an enterprise's competitiveness, such an assumption can avoid the disturbance of price competition. (3) The production costs of enterprises are convex functions with a linearly increasing margin, as shown in Eq. (2).

Cfi = mf qfi2 + cf (mf > 0, cf > 0)

1 γ sfi2 2 fi

(2)

⎧ ∂ Πfi ⎪ ⎪ ∂qfi ⎨ ∂ ⎪ Πfi ⎪ ∂ s fi ⎩

qfi* =

sfi* =

(7)

= [a − (qf 1 + qf 2) − qfi − pe ρfi (1 − θfi )] q⃒ fi =qfi* = 0 qfi = qfi*

= [−γfi sfi + pe ] s⃒ fi = s *fi = 0 sfi = s *fi

(8)

2mf a + pe (ρfj (1 − θfj ) − ρfi (2mf + 1)(1 − θfi )) 2(mf + 1)(2mf + 1)

(9)

1 p γfi e

(10) *

It can be seen from qfi that the proportions of free quotas and carbon intensities will contribute to the output change. The relative size of the two carbon intensities will influence its equilibrium advantage in the product market. The threshold of the carbon intensity ratio is (2mf + 1)(1 − θfi) 1 − θfj

. Thus, when ρfj >

ρfi (2mf + 1)(1 − θfi) 1 − θfj

, (i ≠ j ) , the enterprise

with lower carbon intensity will take this advantage to expand its market share; while when ρfj <

ρfi (2mf + 1)(1 − θfi) 1 − θfj

, (i ≠ j ) , the carbon in-

tensity between the two enterprises make little difference. At this point, even the one with lower carbon intensity will decline in output after the implementation of ETS. In addition, the optimal abatement is inversely proportional to the marginal abatement cost coefficient γfi; namely, the higher the coefficient, the lower its independent emission reduction. We can also see that the optimal abatement taken by the enterprise is related to the carbon price. It is not until the marginal abatement cost is equal to the carbon price that enterprises change from reducing emissions by themselves to purchasing carbon credits in order to meet their needs. Thus, if the carbon price is very low, an enterprise will lose the incentive to develop research and investment on low carbon technologies, as it can buy carbon credits cheaply. It is not a tough task for them to achieve the reduction target. Based on a similar analysis, we can obtain the total output of the

(3)

(4)

(6) The government will freely allocate quota efi to enterprise fi in accordance with the overall emissions reduction plan or its historical carbon emissions, and expect that the emission of enterprise fi will not surpass this cap [5]. We assume that quota efi is proportionate to the output of this enterprise, namely efi = θfi ρfi qfi . The quota is assumed to be used only in the same year [29].

industry q =

2mf a − mf pe (ρfi (1 − θfi) + ρfj (1 − θfj )) (mf + 1)(2mf + 1)

, which is the sum of the duo-

poly enterprises’ outputs. Under the scenario without carbon quota constraints, the total industry output is 2a . Obviously, the industrial 2mf + 3

output will decrease to some extent after enterprises participate in ETS, indicating that implementation of a cap-and-trade system has a negative impact on the industry's total output. In addition, the reduction in the amount of industrial output depends on the general carbon intensity of the industry and the free quotas allocated to each enterprise. A generous free quota may cause a lack of buyer requirements and result in thin trading, while a tight one might make it rather difficult for companies to adapt to the constraints of carbon quota, leading to a

The quotas can be sold on the carbon market. During the process, the amount of quota purchased (sold) by enterprise fi is:

ΔEfi = Efi − efi = qfi ρfi − sfi − efi

1 γ sfi2 − pe (qfi ρfi − sfi − efi ) 2 fi

we can obtain Eq. (9) and Eq. (10):

The net emission that enterprise fi emits is shown in Eq. (4) [8], in which ρfi qfi is the emission that enterprise fi should emit from the total production process, and sfi is the emission reduction by enterprise itself.

Efi = ρfi qfi − sfi

= pf (qf 1, qf 2) qfi − mf qfi2 − cf −

It can be seen from the definition that for the two enterprises in the duopoly model, the optimal output combination (qfi*, qfj*), as well as the independent emission reduction (sfi*, sfj*) is the Nash equilibrium of the game. By computing the first order condition of the profit function with respect to qfi*and sfi* with respect to qfi* and sfi*, namely

Here, m is the marginal production cost coefficient, and cf is the set up cost. The convexity of the cost function is used to illustrate the fact that during production, the operational efficiency of machines and equipment is decreased, but the damage to them increases as the output increases, resulting in an increasing unit production cost [6]. Aiming to focus on the operational decision caused by the carbon quota constraints, the production technologies of these two manufacturers are set to be the same in order to avoid the impact of production costs on the output. Therefore, m and cf are set as constants. (4) The carbon emissions of each enterprise are linearly correlated with the production output [32], which means enterprise fi produces ρfi unit of carbon emissions with one unit of production output. (5) Given that enterprises will manage to reduce their CO2 emissions to some extent, we take individual abatement into consideration. sfi the initial mitigation amount(the emission reduction by taking lowcarbon technologies or by enhancing energy efficiency). The total abatement cost is set according to Eq. (3). γfi is the marginal abatement cost coefficient, the positive value of which means the increment of total abatement cost will increase with the further step of emission abatement. This is because when technologies reach a certain level, the difficulty for further emission reduction will increase.

TACfi =

(6) *

(5)

where ΔEfi < 0 means that enterprise fi has surplus quota to sell, whereas ΔEfi > 0 means that it has to purchase carbon credits. 777

Renewable and Sustainable Energy Reviews 92 (2018) 774–783

C. Wang et al.

Table 2 Nomenclature in the Stackelberg model. Symbol

Definition

pe el ef sl

carbon price free carbon quota amount allocated to supplier l free carbon quota amount allocated to manufacturer f The CO2 reduction amount by independent mitigation behavior of supplier l The CO2 reduction amount by independent mitigation behavior of manufacturer f The product output of the manufacture f product price of supplier before ETS implementation product price of supplier after ETS implementation final product price of manufacturer f carbon intensity of supplier l carbon intensity of manufacturer f Raw material consumption intensity The profit of supplier l The profit of manufacturer f The marginal production cost coefficient of supplier The marginal production cost coefficient of manufacturer Marginal coefficient for enterprise fi The proportion of free quota accounted for actual emissions for enterprise fi production cost of suppler production cost of manufacturer

sf qf pl plc pf ρl ρf η πl πf ml mf γfi θfi Cl Cf

⎧ πl = ηqf (plc , pe )(plc − ml ηqf (plc , pe )) − cl − pe (ρl ηqf (plc , pe ) − el ) ⎨ πf = qf (plc , pe )(pf − ηplc − mf qf (plc , pe )) − cf − pe (ρf qf (plc , pe ) − ef ) ⎩ (12) among which ρl and ρf are carbon intensity of supplier l and manufacturer f, respectively and el and ef are the free carbon quota amounts allocated to supplier l and manufacturer f, respectively. We will not take sl and sf into consideration again because we have already discussed the relationship between the initiative mitigation amount and the carbon price. Here, we calculate the solution when neither the supplier nor the manufacturer has a carbon quota constraint. We then analyze the cost and profit changes for each player when they both have carbon quota constraints and when the supplier can pass part of the cost increment on to its own product. In particular for downstream enterprise, the structure of the profit change will be decomposed to explore potential countermeasures. 3.2.1. Nash equilibrium solutions without ETS In the Stackelberg model, supplier l will determine pl to maximize πl (pl , qf* (pl )) of Eq. (12) and the manufacturer determine the optimal qf* (pl ) to maximize πf (pl , qf (pl )) of Eq. (12). By computing the firstorder condition of Eq. (12), we can obtain:

qf* (pl ) =

potential boycott, especially in the initial stage of establishing ETS pilots.

(13)

The supplier l can expect such a strategy of manufacturer f and will therefore determine the optimal pl* to maximize πl (pl , qf* (pl )) . Invoking the first order condition, we can get:

3.2. The dynamics of the supply partnership caused by indirect cost

pl* =

In this section, a Stackelberg model is applied to discuss the influence of indirect cost derived from ETS, mainly focusing on the supply partnership of the downstream manufacturer. In a supply chain with one manufacturer and one supplier, we endowed the supplier with pricing power over their product as the leader l, and the manufacturer has to use the upstream product as raw material and take the material price as the follower f. The variables and parameters used in the Stackelberg model are explained in Table 2. We have identified the production cost for the manufacturer. Similarly, the production cost for the supplier can be assumed to be Cl = ml ql2 + cl = ml ηqf2 + cl, ml > 0, cl > 0 . As the leader, supplier l has pricing power over the material used in the final product. The supplier will consider its production cost Cl and decide the price pl to maximize profit πl. Moreover, as a follower, after observing the pricing behavior of the upstream supplier, manufacturer f will decide the optimal output qf in order to maximize its own profit according to the cost schedule Cf and material price pl. Before the implementation of ETS, the profit function of each player is:

mf + ml η2 η (2mf + ml η2)

pf

(14)

Therefore, we have qf* (pl* ) =

1 p, 2(2mf + ml η2) f

which means when

aiming to achieve maximum profit, the supplier determines its product's price (the material for the final product) as pl*, and the manufacturer will determine its output as qf* (pl* ) . Their profits are then given by:

1 ⎧π (p * , q* (p * )) = p2 − cl l l f l ⎨ 4(2mf + ml η2) f ⎩ mf πf (pl* , qf* (pl* )) = p 2 − cf 4(2mf + ml η2) f

(15)

3.2.2. Nash equilibrium solutions under ETS When both the supplier and manufacturer work under carbon quota constraints, the production cost of the supplier will increase, resulting in the higher price plc of the raw material used in the final product. Applying a similar analysis process to that used in the scenario without ETS, the manufacturer in the second phase will determine its optimal output qf (plc , pe ) , after considering plc and pe to maximize its profit. Thus, we obtain:

{πl (pl , qf (pl )) = ηqf (pl )(pl − ml ηqf (pl )) − cl πf (pl , qf (pl )) = qf (pl )(pf − ηpl − mf qf (pl )) − cf

1 (p − ηpl ) 2mf f

(11)

qf* (plc , pe ) =

where, pf is the price of the final product produced by the downstream manufacturer and η is the raw material consumption intensity, i.e., the raw material consumption per unit of the final product during the production process. After the implementation of carbon emission trading scheme, both the supplier and the manufacturer will be given carbon quota constraints. The supplier will pass part of the cost increment through to its product, but the manufacturer cannot raise the final product price despite facing higher raw material prices. It merely maintains the original price. At this time, the manufacturer will consider the new material price, carbon price, and mitigation cost in order to balance the total emission reduction and production output. The profits of the enterprises are as follows:

pf − ηplc − pe (1 − θf ) ρf 2mf

(16) *

As supplier l determines the optimal plc to maximize profit by expecting such a strategy from manufacturer f:

plc* = pl* +

mf ηρl (1 − θl ) − (mf + ml η2) ρf (1 − θf ) η (2mf + ml η2)

pe

(17)

It can be found that when the manufacturer can only maintain the original price of its final product to maximize profit, the supplier must consider both its own and the manufacturer's cost schedules and carbon intensities as well as the raw material consumption intensity and the carbon price to determine how to price its product, i.e., whether it is necessary to raise the price and how much the increment ought to be. 778

Renewable and Sustainable Energy Reviews 92 (2018) 774–783

C. Wang et al.

When

ρl ρf

≤

(mf + ml η2)(1 − θf ) mf η (1 − θl)

motivate enterprises to develop low-carbon production. The profit of the manufacturer under ETS is presented in Eq. (21), in which the second item, − ηqf* (pl* ) Δpl , is the profit loss caused by the raw material price increment, which can be considered as the negative impact of the indirect cost increment brought about by the carbon quota constraints. The third item, − Δqf pf , is the profit loss caused by the reduction in output. The fourth item, − pe (ρf qf* (plc, pe ) − ef ), is the profit loss caused by the introduction of a carbon price, which can be considered as the impact of the direct cost increment brought about by the introduction of carbon quota constraints. It should be noted that for enterprises, the fourth item can be changed into a profit increase instead of a loss thereof. If enterprises can limit their CO2 emissions under a free quota, they can earn some profit by selling any surplus CO2 emissions quota. In fact, there are many ways for Chinese enterprises to reduce their CO2 emissions, such as reducing their CO2 intensity by enhancing energy efficiency or by independently mitigating such emissions through clean production. A carbon emission offsetting mechanism is also a good way to offset some of their CO2 emissions. The seven carbon emission trading pilots that have been established so far in China mainly allocate carbon quotas freely. There is thus an ideal time window for Chinese enterprises to take initiative in reducing CO2 emissions. They can take carbon emissions trading as an investment opportunity designed to increase their profits. For governments, however, the free quota amount allocated to enterprises should be fully considered. Such a system will cause much damage to enterprises’ profits and be harmful for enterprises’ development if the free quota is too small; however, if, for each enterprise, the free quota amount represents a surplus, the emission trading will lose the force regarding restraints that it should otherwise have. The fifth term, Δqf (pf − pe ρf ) , can be considered to affect the dynamics between the final product price and the carbon price. When the price of a firm's final product is less than its carbon cost, i.e., pf < pe ρf , the fifth term is negative, indicating that the profit of the manufacturers decreases. This result also demonstrates that for some industries, such as the PV industry, excessive competition over price can only further damage profits, under the circumstance of carbon quota constraints.

, there is plc* < pl*, indicating that in the case

where the carbon quota constraint damages the production capacity of the manufacturer with very high carbon intensities so drastically that it will, in turn, reduce the profit of upstream supplier. To maximize its own profit, the upstream supplier will reduce its price in order to ensure that its sales can still be supported by the demand from the downstream manufacturer. When

ρl ρf

>

(mf + ml η2)(1 − θf ) mf η (1 − θl)

, to earn maximum profit, the

supplier will raise its prices to some extent. The increment will be:

Δpl =

mf ηρl (1 − θl ) − (mf + ml η2) ρf (1 − θf ) η (2mf + ml η2)

pe

(18)

For the manufacturer, its cost increment comes from the carbon price, the material price increment, and its independent mitigation behavior. The cost increment per unit product is

Δcf = pe ρf (1 − θf ) +

pe2 2qf

+ ηΔpl .

Given

plc* = pl* + Δpl ,

comparing

qf* (plc , pe ) with qf* (pl , pe ) , we find that the change of optimal output is as follows: qf* (plc , pe ) = qf* (pl , pe ) − =

pe ρf (1 − θf ) + ηΔpl 2mf

pf − pe (ρf (1 − θf ) + ηρl (1 − θl )) 2(2mf + ml η2)

(19)

Thus, after the implementation of ETS, the output decrement is:

Δqf =

pe (ρf (1 − θf ) + ηρl (1 − θl )) 2(2mf + ml η2)

(20)

The profit of the manufacturer will be:

πfc = πf − ηqf* (pl* ) Δpl − Δqf pf − pe (ρf qf* (plc, pe ) − ef ) + Δqf (pf − pe ρf ) (21) 3.2.3. The analysis and discussion regarding the dynamics of the supply partnership with carbon constraints When both the supplier and manufacturer are participants in the ETS, the supplier can pass part of the cost increment onto its product. However, the manufacturer will not raise the price of the final product, which means that the final product price will remain unchanged. After the implementation of ETS, the output reduction of the manufacturer will be according to Eq. (20). It can be seen that when facing the double pressure of carbon quota constraints and a raw material price increment, the manufacturer must reduce its output accordingly in order to maximize profit. Noting that this output reduction is related to the carbon intensities of both the supplier and manufacturer, it would help to cushion the reduction in profit to some extent if the manufacturer can reduce its carbon intensity. The carbon intensity can affect not only its changes in output but also the mutual interest dynamics between upstream and downstream enterprises across the whole supply chain. As mentioned, when ρl (1 − θl) m + m η2 ≤ 2 1 , we have plc* < pl*, indicating that the supplier will ρf (1 − θf )

4. Analysis of a case study To investigate the application of this model and examine the different impacts of costs derived from carbon quota constraints, a case study is designed for analysis. Hubei Province is one of China's ETS pilots. Administered by Hubei government, Huangshi City has two manufacturers for aluminium alloy section and both of them are incorporated in the ETS. It should be noted that the enterprises in our model are duopoly, however, the two manufacturers are not duopoly enterprises from the national level. But considering that the local demand for aluminium alloy section is mainly satisfied by these two enterprises and they only take part in the local ETS, the geographical scope of the case can be limited to within the Huangshi City. The market demand is assumed to be met all by the two manufacturers and there is no import outside the city. Out of non-disclosure agreement, the two manufacturers are named as M and Z, respectively. They use electricity power as the main energy to produce aluminium alloy section. Some of processing machines, such as melting furnace, heating furnace for aluminium rod, and curing furnace burn natural gas. Manufacturer M didn’t upgrade its equipment until 2014, therefore, some bituminous coal was combusted in 2013. According to energy consumption and output data (from 2013 to 2015) provided by enterprises, we calculate their CO2 emissions (Table 3). In 2013, the carbon intensities of M and Z were 1.77 tCO2/t and 0.85 tCO2/t, respectively. Then, through updating the equipment, the carbon intensity of manufacturer M fell down to 0.97 tCO2/t in 2015. According to industrial report of aluminium product, the unit producing cost is assumed as 1000 yuan/t, and the production cost follows

m2 η

reduce the price of its product in order to ensure demand from the downstream manufacturer. In fact, what is more likely to happen is that in this case, the upstream supplier will change its downstream partner and cooperate with the one with lower carbon intensity in order to enhance profitability. For the manufacturer, such a change in dynamics can be treated as a process of survival of the fittest. By solving this model, we have shown the theoretical importance of reducing the carbon intensity for enterprises. It has always been a critical direction for China's low carbon development to raise energy efficiency and reduce carbon intensity. Chinese enterprises, however, have not fully understood the significance of low-carbon technology and they have therefore been reluctant to invest in this aspect until its benefits are better understood. The implementation of carbon emissions trading will 779

Renewable and Sustainable Energy Reviews 92 (2018) 774–783

C. Wang et al.

Table 3 The CO2 emissions of the aluminium alloy sector manufacturers in Huangshi City. Manufacturer M

2013 2014 2015

Manufacturer Z

CO2 emissions (ton)

Output (ton)

Carbon intensity (tCO2/t)

CO2 emissions (ton)

Output (ton)

Carbon intensity (tCO2/t)

48,926 35,655 41,469

27,859.12 36,351.61 42,934.52

1.77 0.98 0.97

28,051 26,124 26,313

33,040 32,142 36,008

0.85 0.81 0.73

Cf = 0.03qf2 + 3000 . The inverse demand function is set as P (qm , qz ) = 91000 − (qm + qz ) . The marginal abatement coefficients are set according to the characteristics of abatement cost (Table 4). The production cost of upstream supplier follows Cl = 0.01ql2 + 3000 . At the same time, because free quota e should smaller than total emissions from two enterprises before ETS implementation, we set e varies between (30,000, 70,000) to analyze the sensitivity of output change to free quota amount. Based on the data from Chinese seven ETS pilots, we find the current carbon price floats between 20 and 80 Yuan/ton. Therefore, we assume that the carbon price is 50Yuan/ton. The carbon quota constraints make carbon intensity as a competitiveness of the enterprise. For the enterprises within the same industry, the one with less carbon intensity will enlarge its market proportion by taking the advantage of its carbon intensity. The output changes in year 2013 and 2015 are shown in Fig. 1 and Fig. 2, respectively. In 2013, manufacturer Z had lower carbon intensity and stronger carbon competitiveness. From Fig. 1 we can see that although carbon quota constraints improved the cost, the manufacturer Z could use its carbon intensity advantage to enlarge its product output. While for the manufacturer M, the situation was not that good. On one hand, the carbon constraints exerted carbon cost pressure on it; on the other hand, the shortage of its carbon intensity made it lack of carbon competitiveness, leading to much product loss. What's worse, with tighter free quota amount, the product loss would become larger. Accordingly, compared to Z, the output gap would become larger. Before the implementation of ETS, the duopoly enterprises had almost the same market share. That situation was broken because of the carbon quota constraints and the manufacturer Z had advantages to enlarge its market share. When the free quota is 65,000, it is similar as the reality that the free quota covers 95% of the actual emissions. According to model results, the optimal output of M and Z in 2013 should be about 24,834ton, and 30,314ton, respectively. Compared to actual outputs, 27,859 and 33,040, the differences are very small and the trends of output changes are consistent with the model, by which the model is verified. But by upgrading its processing equipment, replacing the bituminous coal combustion machines with natural gas combustion ones, the carbon intensity of manufacturer M fell down a lot in 2015. As a result, the carbon intensity gap between the two manufacturers became smaller although the carbon intensity of M was still larger. Fig. 2 shows that Z would still have advantages on product market, however, compared to non-ETS situation, its output would also reduce. Compared to

Fig. 1. The output changes of two manufacturers in year 2013.

Fig. 2. The output changes of two manufacturers in year 2015.

actual outputs in 2015, the model results are smaller. The reason is that we assume the demand function in 2015 the same as in 2013, however, the demand should be more because of economy development. But this assumption does not affect the trends of output changes. That is to say, enterprise with lower carbon intensity has larger market share. And with tighter quota amount, the advantage becomes more obvious. From the perspective of supply chain, the carbon intensity will impact not only product output but also the partner partnership between upstream and downstream enterprises. We take manufacturer M as an example to discuss the supply partnership change. The indirect cost derived from ETS makes manufacturer M change its product output strategy and supplier's pricing strategy. When

ρl ρf

>

(mf + ml η2)(1 − θf ) mf η (1 − θl)

, the

supplier will raise the price of its product for transmitting the carbon cost to product price, leading to the indirect cost pressure for downstream enterprise. By using δ to represent ρ (1 − θ) , we can see that with the carbon cost increasing, the manufacturer will reduce the output to meet the carbon constraint target. According to Eq. (20), assuming that producing one unit of downstream product consumes one unit of upstream product, namely η = 1, the impact of indirect cost on output reduction of downstream enterprise is shown in Fig. 3.

Table 4 The corresponding parameters of duopoly enterprises in the case. larger intensity difference Enterprise M Carbon intensity(ton/unit) Marginal abatement cost coefficients(yuan/tCO2) Inverse demand function

Smaller intensity difference Enterprise Z

1.77 0.85 2 5 P (qm , qz ) = 91000 − (qm + qz )

780

Enterprise M

Enterprise Z

0.97 4

0.73 7

Renewable and Sustainable Energy Reviews 92 (2018) 774–783

C. Wang et al.

very high, this enterprise will greatly reduce the output for meeting the emission reduction target, leading to a decreasing demand for the upstream product. Thus, the supplier has to reduce the price of its product to some extent in order to prevent the demand from dropping too much. The implementation of ETS may break the original supply partnership and incentivize a new one to form. 5. Inspiration To achieve the CO2 reduction target, in light of the advanced experiences of developed countries with their own carbon emission trading systems, the Chinese Government has floated the proposal “… establishing a proper statistical and accounting system for CO2 emissions and gradually establishing a carbon emissions trading market …” in the 12thfive-year plan. It is the first time that the Central Government has proposed the establishment of a domestic carbon market for China in such an official document, indicating that this issue has been on the agenda a while. The carbon market, on one hand, will raise energy prices, which in turn leads to increasing production cost; on the other hand, it will promote potential innovation in mitigation technology by internalizing the externalities of the carbon emissions of a business. Given that individual enterprises often locate themselves within a supply chain network, the carbon quota constraints impact not only their own production behavior, but the supply chain to which they belong. After analyzing the effects of direct and indirect costs derived from ETS, the research offers inspiration for Chinese enterprises at this burgeoning stage of ETS.

Fig. 3. The impact of indirect cost on enterprise M's output reduction.

The line with δl = 0 represents the situation when only the downstream manufacturer has carbon constraints while the upstream supplier not. Such output reduction is caused by purchasing carbon permits, that is, the direct impact of carbon quota constraints. The line with δf = 0 represents the situation when only upstream supplier has carbon quota constraints and passes all the carbon cost through to the price of its product. Such output reduction reflects the impact of indirect cost derived from ETS on downstream manufacturer. But in reality, even if that both upstream supplier and downstream manufacturer have carbon quota constraints, it is impossible for supplier to transmit all the carbon cost to its product. The downstream manufacture also has the wiggle room to bargain for the price of upstream product. The price change of upstream product under ETS is shown in Fig. 4. With δf increasing, the supplier tends to reduce the product price to some extent in order to maintain a higher level of demand and maximize the profit. When downstream manufacturer does not participate in ETS, the supplier is more likely to raise its product price. It indicates that supplier prefers the downstream manufacturer with lower carbon intensity as their partner because there is no need to reduce the price to rescue the sharply decreasing demand. To sum up, it is very important for enterprises to reduce their carbon intensity, especially at the stage of ETS pilots in China. On one hand, from the perspective of direct cost, a lower carbon intensity will lead to a higher carbon competitiveness. The enterprise with lower carbon intensity can take this advantage to enlarge the market share, while the output of the one with higher carbon intensity will be hit by the increasing cost for purchasing carbon permits. On the other hand, from the perspective of indirect cost, the supplier prefers to choose the downstream manufacturer with lower carbon intensity as its partner. The reason is that if the carbon intensity of downstream manufacturer is

(1) Enterprises have to stop their wait-to-see attitude and do their best to reduce their carbon intensities. We have made some investigation with some officials in China Hubei Emissions Exchange(CHEEX) and Center of Hubei Cooperative Innovation For Emissions Trading System. What they said proved that there are still some enterprises having not pay enough attention to the challenge of ETS. However, some companies have already obtained profits from selling the carbon quotas. Hubei formally started their emission trading activity in 2014. During that trading year, by our investigation, we get an idea that the first permits seller is Hubei Energy Group. Before launching ETS, they had just finished their re-construction of energy system which had greatly improved their energy efficiency and reducing the carbon intensity. Thus, when the free quota was allocated, they argue they would achieve the target easily and have some surplus quotas for sure. This is an example that carbon quotas may bring profits to companies. Through the investigation, we also found that in Hubei's first trading year, most companies just waited and bought extra allowances at the end of the trading year, at higher prices of course. These companies now show great interests in carbon intensity reduction and carbon management. The investment in carbon management has already been stimulated among the enterprises who have already involved in ETS. However, among enterprises excluded from ETS currently, less attention is paid to this field. With the development of pilots, more industries are absorbed in ETS year by year. It is an important path to make preparation in advance in order to get the carbon competitiveness. In the first trading year, some enterprises in the iron & steel industry also achieve the reduction target easily, the most reduction, however, came from output reduction not carbon intensity reduction. It should be noted that the output reduction is not a result of carbon pressure. It was caused by economic situation of the whole industry. With the carbon quota constraints becoming stricter, the output reduction cannot be the measure to deal with reduction targets. From long-term perspective, the energy efficiency has to be enhanced gradually. (2) The implementation of ETS is an opportunity to break traditional supply partnership and form low carbon one. Our research indicates

Fig. 4. The price change of upstream supplier's product. 781

Renewable and Sustainable Energy Reviews 92 (2018) 774–783

C. Wang et al.

Acknowledgement

that a higher carbon intensity will not only hit an enterprise's output, but also probably cause its supplier to pursue a new partnership with lower carbon intensity. This may hurt the profits of some enterprises, but for the development of China's whole economy, it has greatly progressive significance. (3) Carbon quota constraints on the whole supply chain can motivate enterprises to take initiatives in disclosing their production and mitigation data. Information transparency can help enterprises not only to determine optimal output strategy, but also to choose proper upstream or downstream partners, and obtain the best benefits therefrom. (4) Carbon quota constraints motivate enterprises to take initiatives in reducing CO2 emissions. ETS links enterprises’ profits directly with CO2 emissions, which has become an important issue that enterprises need to carefully consider. Enterprises can control their emissions within the free quota coverage by vigorously participating in carbon offsetting projects. China's ETS is still in its infancy currently, which is an opportunity for enterprises to better understand and adapt to it.

This study is supported by the National Science Fund for Distinguished Young Scholars (Reference No. 71625003), National Key Research and Development Program of China (Reference No. 2016YFA0602504), National Natural Science Foundation of China (Reference No. 91746208, 71173017, 71573016, 71521002), Beijing Social Science Foundation (Reference No. 17JDGLA010) and Fundamental Research Funds for the Central Universities (Reference No. FRF-TP-17-015A1). The authors want to thank Professor Yiming Wei and Dr. Bin Zhang for their comments and suggestions. References [1] Abdallah T, Farhat A, Diabat A, Kennedy S. Green supply chains with carbon trading and environmental sourcing: formulation and life cycle assessment. Appl Math Model 2012;36:4271–85. [2] Chaabane A, Ramudhin A, Paquet M. Design of sustainable supply chains under the emission trading scheme. Int J Prod Econ 2012;135:37–49. [3] Chan HS, Li SJ, Zhang F. Firm competitiveness and the European Union emissions trading scheme. Energy Policy 2013;63:1056–64. [4] Chen YH, Sijm J, Hobbs BF, Lise W. Implications of CO(2) emissions trading for short-run electricity market outcomes in northwest Europe. J Regul Econ 2008;34:251–81. [5] Du S, Zhu L, Liang L, Ma F. Emission-dependent supply chain and environmentpolicy-making in the ‘cap-and-trade’ system. Energy Policy 2013;57:61–7. [6] Fan J-L, Ke R-Y, Yu S, Wei Y-M. How does coal-electricity price linkage impact on the profit of enterprises in China? Evidence from a Stackelberg game model. Resour Conserv Recycl 2018;129:383–91. [7] Huang Y, Wang K, Zhang T, Pang C. Green supply chain coordination with greenhouse gases emissions management: a game-theoretic approach. J Clean Prod 2016;112:2004–14. [8] Jones R, Mendelson H. Information goods vs. industrial goods: cost structure and competition. Manag Sci 2011;57:1100–262. [9] IPCC (Intergovernmental Panel on Climate Change). Climate change 2007: Synthesis Report. Summary for Policy Makers[EB/OL]. [2007-03-12]; 2007. [10] Kara M, Syri S, Lehtila A, Helynen S, Kekkonen V, Ruska M, Forsstrom J. The impacts of EU CO2 emissions trading on electricity markets and electricity consumers in Finland. Energy Econ 2008;30:193–211. [11] Klaassen G, Nentjes A, Smith M. Testing the theory of emissions trading: experimental evidence on alternative mechanisms for global carbon trading. Ecol Econ 2005;53:47–58. [12] Kuik O, Mulder M. Emissions trading and competitiveness: pros and cons of relative and absolute schemes. Energy Policy 2004;32:737–45. [13] Lund P. Impacts of EU carbon emission trade directive on energy-intensive industries - Indicative micro-economic analyses. Ecol Econ 2007;63:799–806. [14] Meleo L. On the determinants of industrial competitiveness: the European Union emission trading scheme and the Italian paper industry. Energy Policy 2014;74:535–46. [15] Oberndorfer U, Rennings K. Costs and competitiveness effects of the european union emissions trading scheme. Eur Environ 2007;17(1):1–17. [16] Park T, Kim C, Kim H. A real option-based model to valuate CDM projects under uncertain energy policies for emission trading. Appl Energy 2014;131:288–96. [17] Perdan S, Azapagic A. Carbon trading: current schemes and future developments. Energy Policy 2011;39:6040–54. [18] Ponssard JP, Walker N. EU emissions trading and the cement sector: a spatial competition analysis. Clim Policy 2008;8:467–93. [19] Ren J, Bian Y, Xu X, He P. Allocation of product-related carbon emission abatement target in a make-to-order supply chain. Comput Ind Eng 2015;80:181–94. [20] Rietbergen MG, Blok K. Assessing the potential impact of the CO2 Performance Ladder on the reduction of carbon dioxide emissions in the Netherlands. J Clean Prod 2013;52:33–45. [21] Sartzetakis ES. On the efficiency of competitive markets for emission permits. Environ Resour Econ 2004;27(1):1–19. [22] Sheu J-B. Bargaining framework for competitive green supply chains under governmental financial intervention. Transp Res Part E: Logist Transp Rev 2011;47:573–92. [23] Sikhar B, Gaurav A, Zhang WJ, Biswajit M, Tiwari MK. A decision framework for the analysis of green supply chain contracts: an evolutionary game approach. Expert Syst Appl 2012;39:2965–76. [24] Skelton A. EU corporate action as a driver for global emissions abatement: a structural analysis of EU international supply chain carbon dioxide emissions. Glob Environ Change 2013;23:1795–806. [25] Smale R, Hartley M, Hepburn C, Ward J, Grubb M. The impact of CO2emissions trading on firm profits and market prices. Clim Policy 2006;6:31–48. [26] Tomas RAF, Ribeiro FR, Santos VMS, Gomes JFP, Bordado JCM. Assessment of the impact of the European CO2 emissions trading scheme on the Portuguese chemical industry. Energy Policy 2010;38:626–32. [27] Tsai MT, Yen CW. The influence of carbon dioxide trading scheme on economic dispatch of generators. Appl Energy 2011;88:4811–6. [28] Wang S, Wan L, Li T, Luo B, Wang C. Exploring the effect of cap-and-trade

Undeniably, the direct and indirect cost increments derived from ETS will be a challenge for enterprises. It is also a rare opportunity in the sense that it motivates enterprises to pay more attention to raise their energy efficiency and reduce their CO2 emissions.

6. Conclusion The carbon emission trading scheme (ETS) has become one of the most popular measures used to achieve CO2 emissions reduction targets in many countries because of its flexibility and cost-effectiveness. By drawing on the experience of ETS in developed countries, the Chinese government has gradually established seven ETS pilots, aimed at exploring ways of establishing a unified carbon market. This research employs a game theory framework to analyze the impacts of direct and indirect costs derived from carbon quota constraints. The direct cost affects the enterprises’ competitiveness and indirect cost affects their supply partnership. Model analysis indicates that after introducing carbon quota constraints, for manufacturers that are located downstream of the supply chains, the costs may increase. The carbon intensity of an enterprise will play a significant role in coping with carbon constraints. On one hand, a lower carbon intensity will help an enterprise to gain an advantage of carbon competitiveness, by which the enterprise can enlarge the market share. Moreover, when carbon quota constraint is tighter, the advantage will become more obvious. Therefore, from the perspective of competitiveness, the enterprise with lower carbon intensity will have more possibility to get benefits. On the other hand, an excessively high carbon intensity will impair output capacity, thus impacting profits. This is similar to the conclusion of the study of Wang et al. [28], which suggests that enterprise's optimal production quantity is positively related to free carbon quotas, while negatively to carbon emission rate. And we further analyzes the effects of carbon intensity under different condition of free carbon quotas. From the perspective of supply partnership, ETS will motivate an upstream supplier to search for a new partner with lower carbon intensity. For China, at this stage, these results bear more profound significance and provide inspiration: at the budding stage of ETS, Enterprises should take advantage of this opportunity to innovate their production and mitigation technologies to reduce their carbon intensity. The empirical research on the effects of ETS on enterprises in China needs to be further studied, which has difficulties in obtaining data from enterprises currently because enterprises are often reluctant to provide operation and environmental data.

782

Renewable and Sustainable Energy Reviews 92 (2018) 774–783

C. Wang et al. mechanism on firm's production planning and emission reduction strategy. J Clean Prod 2018;172:591–601. [29] Wang Z, Wang C. How carbon offsetting scheme impacts the duopoly output in production and abatement: analysis in the context of carbon cap-and-trade. J Clean Prod 2015;103:715–23. [30] Xu J, Chen Y, Bai Q. A two-echelon sustainable supply chain coordination under cap-and-trade regulation. J Clean Prod 2016;135:42–56. [31] Xu J, Qi Q, Bai Q. Coordinating a dual-channel supply chain with price discount contracts under carbon emission capacity regulation. Appl Math Model

2018;56:449–68. [32] Yang L, Zhang Q, Ji J. Pricing and carbon emission reduction decisions in supply chains with vertical and horizontal cooperation. Int J Prod Econ 2017;191:286–97. [33] Zhou P, Zhang L, Zhou DQ, Xia WJ. Modeling economic performance of interprovincial CO2 emission reduction quota trading in China. Appl Energy 2013;112:1518–28. [34] Zu Y, Chen L, Fan Y. Research on low-carbon strategies in supply chain with environmental regulations based on differential game. J Clean Prod 2018;177:527–46.

783