Journal of Cleaner Production 139 (2016) 337e346

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Do carbon taxes work? Analysis of government policies and enterprise

strategies in equilibrium
Tsai Chi Kuo a, I-Hsuan Hong b, *, Sheng Chun Lin a
a
Department of Industrial and Systems Engineering, Chung Yuan Christian University, Taiwan
b
Institute of Industrial Engineering, National Taiwan University, Taiwan

a r t i c l e i n f o a b s t r a c t

Article history: To achieve the goal of carbon emission reduction, a number of countries have implemented carbon tax
Received 8 January 2016 policies, and carbon pricing has become an important tool. This paper discusses the interactions between
Received in revised form a government and an enterprise for implementing carbon taxes over time or by level of tax, considering
6 July 2016
greenhouse gas emission. We examine the impacts of various carbon tax policies on enterprises' in-
Accepted 25 July 2016
Available online 2 August 2016
vestments in new technology, their reductions in production quantity, and governments' tax remittances.
A case study of the strategic responses of three firms in Taiwan's electronics, cement, and steel industries
to two carbon emission policies promulgated by government shows that appropriate levels of tax can
Keywords:
Carbon tax
induce an enterprise to alter its production processes to achieve emission reduction as part of a long-
Carbon emission term business strategy.
Government policy and enterprise strategy © 2016 Elsevier Ltd. All rights reserved.

1. Introduction economic impacts on individual enterprises. A review of the pub-

lished literature shows that most studies focus on the national level
A carbon tax is an excise that deals exclusively with carbon- to explore the impacts of carbon tax on the behaviors of firms and
based fuel consumption (Zhang and Baranzinic, 2004). Some customers. For the national level, the most popular model is
countries in Europe, such as the Netherlands, Sweden, Finland, and computable general equilibrium (CGE), because it is able to capture
Norway, have implemented carbon taxes for years (Baranzini et al., the complicated interplays among economic variables and policy
2000). Policy-makers throughout the world are stepping up efforts instruments. CGE shows both direct and indirect effects of the
to reduce carbon dioxide (CO2) emission and greenhouse gases implementation of a tax, e.g., (micro-economic) changes in demand
(GHG) (Zimmer et al., 2015). For instance, the US EPA has proposed and supply due to price changes and interactions of different
uniform national limits on the amount of carbon pollution that markets at the macroeconomic level (Elliott and Fullerton, 2014).
future power plants will be allowed to emit (EPA, 2013). Nakata and Lamont (2001), who propose a partial equilibrium
Lin and Xie (2016) provide some policy advice for constraining model to study the effects of a carbon tax on energy sources in
and reducing CO2 emission in China's food industry. Gupta (2016) Japan, demonstrate that a well-designed carbon tax system leads to
states that a tax system has direct effects in reducing GHG emis- a reduction in CO2 emission. Scrimgeour et al. (2005) study the
sions. At the national level, the imposition of a carbon tax can effects of a carbon tax on the New Zealand economy by means of a
impact competitiveness, gross domestic product (GDP), and the computable general equilibrium model; their results, however,
export and import of goods (Saddler et al., 2006; Zhang and indicate an unfavorable effect on GDP. Bruvoll and Larsen (2004)
Baranzinic, 2004, Schaltegger and Synnestvedt, 2002; Kim, 2008; find only a 2% decrease in emission after Norway levied a carbon
Lu et al., 2010); at the industrial level, imposition can vary by in- tax in 1991, whereas Johansson (2000) find that certain industries
dustries, e.g., energy (Lee et al., 2007, 2008); and at the firm level either receive exemptions or do not pay the full tax based on
(Huisingh et al., 2015; Baranzini et al., 2000), it can have significant government policy after Sweden introduced a carbon tax in the
same year. Lu et al. (2010) create a dynamic recursive general
equilibrium model to evaluate the effects of carbon taxes in China;
* Corresponding author. Department of Mechanical Engineering, National Taiwan surprisingly, their results show that the taxes have little effect on
University, Taiwan. Tel.: þ886 2 3366 9507; fax: þ886 2 2362 5856. GDP. Drouet et al. (2008) evaluate a Swiss carbon tax scheme by
E-mail addresses: [email protected] (T.C. Kuo), [email protected] (I.-H. Hong),
means of a computable general equilibrium model to reduce
[email protected] (S.C. Lin).

http://dx.doi.org/10.1016/j.jclepro.2016.07.164
0959-6526/© 2016 Elsevier Ltd. All rights reserved.
338 T.C. Kuo et al. / Journal of Cleaner Production 139 (2016) 337e346

emission and predict the possible levels of carbon tax for 2020 and examine the linkages in Indonesia between economic growth, en-
2050. ergy consumption, financial development, trade openness, and CO2
For the firm level, most research proposes the optimization emissions from first quarter 1975 to fourth quarter 2011. Because
model, where the carbon tax can be considered as a parameter for the government and an individual enterprise can be viewed as two
optimization. Huisingh et al. (2015) investigate technical in- independent decision-makers, each looking for its optimum, it is
novations and policy interventions for improved energy efficiency possible to solve for the Nash equilibrium, where neither player one
and carbon emission reduction in a wide diversity of industries. can be better off by a unilateral change in its decision, to analyze the
Baranzini et al. (2000) show that the design of a carbon tax and the strategies adopted by both.
use of the revenues generated may compensate for the negative To our knowledge the behavior of an individual enterprise in a
impacts of imposition. Young and Huang (2012) propose a two- specific industry is rarely discussed in the published literature. This
phase approach in developing GHG reduction strategies to paper is motivated by our recognition that carefully designed
comply with current international trends. Using Taiwan as an environmental policies that stimulate firms' creative responses and
example, Lee et al. (2007) propose a two-stage price model to improve production efficiency can result in enterprise advantages
discuss the effects of carbon taxes on the iron and steel, chemical, (Porter, 1991; Porter and van der Linde, 1995). We examine how
cement, and paper industries. Ockenden and McAloon (2011) find different carbon tax policies implemented by government affect the
that an increase in carbon taxes would reduce the competitiveness business strategies of specific enterprises, specifically, investing in
of Australia's food industry, especially for imported food stuffs. new technology, curtailing production to control CO2 emission, and
Other studies (Porter and van der Linde, 1995; CSGCC, 2000) remitting the carbon tax. The numerical results of our case study
identify several competitive advantages related to imposition, i.e. a corroborate our theory that an inappropriate carbon tax level often
firm can shift the increased cost of the carbon tax to consumers, prompts enterprises to pay carbon taxes rather than reducing
reduce the carbon content of its products, or avoid paying the tax emission.
altogether by improving its production chain. However, competi- The remainder of this paper is organized as follows. In Section 2,
tive losses are often apparent and have a short-term effect, whereas we give the formal definition of the enterprise profit and the gov-
competitive advantages can be more difficult to quantify, because ernment payoff based on the policy designed by a hypothetical
they mainly accrue in the long term (Zhang and Baranzinic, 2004; government and the strategy chosen by a hypothetical enterprise in
Siriwardana et al., 2011; Baranzini et al., 2000). Table 1 summa- different industries. In Section 3, we present case studies and the
rizes the published studies of mathematical models for a carbon comparative statistics under different scenarios adopted and cho-
tax. sen by the government and enterprises. Section 4 gives our con-
As government and carbon-producers debate additional mech- clusions and suggests some directions for future research.
anisms to control global GHG, it will force enterprises to refashion
their supply chain networks (Chaabane et al., 2012). Hua et al. 2. Research method and procedures
(2011) investigate how firms manage their carbon footprints in
inventory management under a carbon emission trading mecha- This paper uses a hypothetical government's carbon tax policies
nism. The authors analytically derive the optimal order quantity and several hypothetical enterprises' corresponding strategies as
and numerically examine the impacts of carbon trade, carbon price, two competitive factors. The government and the enterprises are
and carbon cap on order decisions, carbon emission, and total cost. two parties in the game, wherein one party's response varies with
Finally, a handful of studies simultaneously consider the in- the other's decision. The goal of the government is to optimize the
teractions between a government and an enterprise regarding the effect of carbon taxes so that the carbon emission can be efficiently
imposition of a carbon tax. Hwang (2011) finds that the limited reduced, and the goal of the enterprise is to optimize its profits. The
economic scale is not appealing enough to attract leading com- government has two possible tax policies: year interval carbon tax
panies to invest capital in clean technology. Shahbaz et al. (2013) price (SA), and price dependent carbon emission level (SB). An

Table 1
Prior studies of mathematical models for a carbon tax.

Model Author Country Issues

Global models of computable (García Chile What is the minimum carbon tax rate? What would be the change in the country's GDP? Which economic
general equilibrium (CGE) Benavente, sectors would be most and least affected by the carbon tax?
2016)
(Allan et al., Scotland It used an empirical energyeeconomyeenvironmental model to simulate the impact of a specific carbon
2014) tax on the levels of carbon emissions and sectoral economic activity.
(Guo et al., China It restructured the energy industry sectors to obtain robust simulation results by using a CGE model.
2014)
(Ruamsuke Southeast It constructed a bottom-up module for energy generation sector that towards 2050.
et al., 2015) Asia
(Meng et al., Australia It simulated the effects of a carbon tax on the environment and the economy.
2013)
Partial equilibrium (Datta, 2010) India It found a fuel-related tax would be as progressive as a carbon tax.
(Nakata and Japan It developed a partial equilibrium model to forecast changes in the energy system in 2040.
Lamont, 2001)
Dynamic programming model (Ma et al., 2016) Not How would a carbon tax affect the optimal order quantity for a manufacturer? How should a government
available implement an effective range of the carbon tax? How should a manufacturer select appropriate suppliers
to satisfy production demand?
Bi-objective optimization model (Fahimnia et al., Not The research achieved two objectives: (1) minimization of overall supply chain costs; (2) minimization of
2015) available total production and transportation carbon equivalent emissions.
Multi-objective optimization (Liu et al., 2016) China A multi-objective optimization model was constructed for cutting parameters for high-efficiency and
low-carbon manufacturing to set the carbon emission and working hour as optimization objectives and to
use the cutting feed rate as the control variable.
 T.C. Kuo et al. / Journal of Cleaner Production 139 (2016) 337e346 339

enterprise has three possible strategic responses: production pro- Table 3

cess flow or equipment improvement (SX), production quantity Enterprise profit and government payoff.

reduction (SY), and no action (SZ). Table 2 lists the details. Government
This paper uses the following notations: SA SB

Enterprise SX ðEPSA SX ; GTSA SX Þ ðEPSB SX ; GTSB SX Þ

Y: the number of year intervals determined by the government
SY ðEPSA SY ; GTSA SY Þ ðEPSB SY ; GTSB SY Þ
NY: the number of years within a year interval SZ ðEPSA SZ ; GTSA SZ Þ ðEPSB SZ ; GTSB SZ Þ
CPYt: the unit carbon tax for year t, where CPY0 ¼ 0
L: the number of levels based on the quantity of carbon emission
NL: the amount of carbon emission per level
enterprise may simply choose paying the carbon tax without
CPLi: the unit carbon tax for level i, where CPL0 ¼ 0, CP0 ¼ 0,
making any attempts to reduce the carbon emission (SZ). The
CP0 ¼ 0]
government payoff and enterprise profit are represented in the bi-
Q: the total selling quantity of products per year
matrix as shown in Table 3, where the first/second argument in the
P: the unit price of products
pair ($,$) is the enterprise profit/government payoff under a specific
C: the total carbon emission per year
combination of the government's adopted policy and the enter-
S: the planning horizon (years)
prise's chosen strategy. For example, GTSA SX and EPSA SX are the
CD: the annual carbon reduction ratio
government payoff and the enterprise profit, respectively, under a
M: the expense for improving production processes or replacing
scenario combination of SA and SX.
equipment
We look for the Nash equilibrium pair of the government policy
MD: the carbon reduction efficiency (0  MD  1)
and enterprise strategy, where both parties have no incentives to
I: the bank interest rate
unilaterally deviate from their chosen policy and strategy. A con-
YS: the ending year of production reduction in order to reduce
ventional way to solve for the Nash equilibrium of this bi-matrix
the carbon emission
game is to determine the government's best response to each of
MR: the equipment depreciation rate (0  MR  1)
the three enterprise strategies (SX, SY, and SZ) by underlining the
Es: the price elasticity coefficient
payoff to the government. Similarly, for each of the government
EPij: the profit of the enterprise when the enterprise adopts
policies (SA and SB), the enterprise chooses its best response to each
strategy i and the government chooses policy j
of the government's policies by underlining the profit to the en-
GTij: the payoff of the government when the enterprise adopts
terprise's best response to each of government's feasible policies. A
strategy i and the government chooses policy j
pair of the policy and strategy satisfies the condition of the Nash
equilibrium if the government's policy is a best response to the
enterprise's strategy and vice versa, i.e., if both the enterprise profit
2.1. Government policies and enterprise strategies and the government payoff are underlined in the corresponding
cell of the bi-matrix. The next section describes the six possible
The government has two possible policies available and the pairs of the government policies and enterprise strategies.
enterprise has three potential strategies to choose. Two possible tax
instruments that the government may adopt are considered. In 2.2. The payoffs of the enterprise and government
Scenario A (SA), the government charges the enterprise an
increasing rate of carbon taxes over year intervals. In Scenario B The enterprise profit is the total sales revenue minus the asso-
(SB), the government charges the enterprise different carbon tax ciated costs corresponding to the strategies adopted by the enter-
rates depending on the total amount of CO2 it generates during the prise. The metric to measure the payoff of the government is the
production process. Under these carbon tax policies, the enterprise decrease in the total carbon emission due to a unit of carbon taxes
can choose to curtail the carbon emission by improving the pro- collected by the government. It implies the government prefers the
duction process (SX) or by cutting down the production quantity most efficient carbon tax policy such that a unit of collected carbon
(SY). In Scenario X (SX), the reduction goal of the annual carbon taxes returns the highest amount of reduction in carbon emission.
emission can be achieved via an improvement of the production
process while the total annual selling quantity remains at the same 2.2.1. SA versus SX
level. Scenario Y (SY) implies that the enterprise may be reluctant to In the pair of SA and SX, the enterprise reduces the carbon
invest in an improvement of the production process to reduce the emission by improving its production process flow or replacing
annual carbon emission, but the enterprise decreases the carbon production equipment (SX), and the government adopts the policy
emission by a low level of production quantity. However, the that the unit carbon tax increases over time in the planning horizon

Table 2
Government tax policies and enterprise's strategic responses.

Government
Scenario A (SA) Year interval carbon tax price
Carbon taxes differ based on the different year interval. (An enterprise pays an increasing rate of carbon taxes over year intervals.)
Scenario B (SB) Price dependent carbon emission level
Carbon taxes differ based on the carbon emission level. (An enterprise with a high level of carbon emission pays a high carbon tax.)
Enterprise
Scenario X (SX) Production process flow or equipment improvement
The enterprise reduces carbon emission by improving production process flow or replacing production equipment.
Scenario Y (SY) Production quantity reduction
The enterprise reduces the production quantity to decrease carbon emission.
Scenario Z (SZ) No action
The enterprise pays the carbon tax, rather than reducing emission.
340 T.C. Kuo et al. / Journal of Cleaner Production 139 (2016) 337e346

(SA). The planning horizon may include several complete year in-
tervals, where each year interval consists of NY years; the years not V1 ¼ 0 and V2 ¼ 1 ; if a ¼ 0
included in a complete year interval are called remaining years in  
an incomplete year interval. The enterprise profit and the govern- S
V1 ¼ 1 and V2 ¼ 0 ; if as0 and ¼0
ment payoff can be stated as NY

8 9 8 9
X
S a <
X X
i$NY 1  = < S1 
X =
Enterprise profitðEPsA sX Þ ¼ PF  M ð1 þ IÞt  V1 C $C 1  MD $MRg  V2 $ CPY b $C 1  MD $MRk
: PYi ; : ;
t¼1 i¼1 g¼ði1Þ$NY k¼a$NY

( ( ) ( ))
Pa Pi$NY 1   PS1  
C$S  V1 i¼1 C g¼ði1Þ$NY
1  MD $MRg þ V2 C k¼a$NY 1  MD $MR
k

Government payoffðGTsA sX Þ ¼ ( ) ( )
Pa Pi$NY 1   PS1  
V1 i¼1 CPYi $C g¼ði1Þ$NY
1  MD $MRg þ V2 CPY b $C k¼a$NY 1  MD $MRk

where  
S
V1 ¼ 1 and V2 ¼ 1 ; if as0 and s0:
PF ¼ S$Q$P: the total sales revenue NY
P
M St¼1 ð1 þ IÞt : the capital investment of facility or process
improvement
P P Y 1 g
V1 ai¼1 fCPYi $C i$N
g¼ði1ÞNY
ð1  MD $MR Þg: the cost of carbon
taxes incurred in complete year intervals 2.2.2. SA versus SY
P
V2 $CPY b $C S1 k
k¼a$NY ð1  MD $MR Þg : the cost of carbon taxes In the pair of SA and SY, the enterprise reduces the carbon
incurred in remaining years in an incomplete year interval. emission by reducing its production quantity (SY), and the gov-
ernment adopts the policy that the unit carbon tax increases over
To identify a and b, we have time in the planning horizon (SA). The enterprise profit and the
government payoff can be stated as

a < 8 9 8 9 n o
  X XY
i$N = < X
YS =
Enterprise profit EPSA SY ¼ PF  V1 C $C
: PYi
ðCD Þf ;  :CPY b $C ðCD Þf ;  CPY b $C$g$ðCD ÞYS
i¼1 f ¼ði1ÞNY þ1 f ¼ða$NY Þþ1
d n
X o n o
 V2 CPYi $C$NY $ðCD ÞYS  V3 $ CPYε $CðS  d$NY ÞðCD ÞYS
i¼bþ1

 
Government payoff GTSA SY ¼
8
< X a 8< XY
i$N 9
=
8
< X
YS 9
= n o d n
X o
C$S  V1 C ðCD Þf ; þ :C ðCD Þf ; þ C$g$ðCD ÞYS þ V2 C$NY $ðCD ÞYS
: :
i¼1 f ¼ði1Þ$NY þ1 f ¼ða$NY Þþ1 i¼bþ1
)
n o
YS
þV3 CðS  d$NY ÞðCD Þ

Pa (
Pi$NY ) (
PYS ) n o P n o
V1 i¼1 CPYi $C f ¼ði1Þ$NY þ1
ðCD Þf þ CPY b $C f ¼ða$NY Þþ1
ðCD Þf þ CPY b $C$g$ðCD ÞYS þ V2 di¼bþ1 CPYi $C$NY $ðCD ÞYS
n o
þV3 $ CPYε $C$ðS  d$NY Þ$ðCD ÞYS

where
 
S S P S
a¼ and b ¼ a þ 1; if <Y PF ¼ Q $P Yg¼1 fðCD Þg ð1 þ ðð1  ðCD Þg Þ=Es ÞÞg þ Q ðCD ÞYS $Pð1þ
NY NY
ðð1  ðCD ÞYS Þ=Es ÞðS  YS Þ: the total sales revenue
P P Y
V1 ai¼1 fCPYi $C i$N ðC Þf g: the cost of carbon tax
f ¼ði1Þ$NY þ1 D
S
a ¼ Y and b ¼ a ; if  Y: incurred in complete year intervals during the periods of pro-
NY duction quantity reduction
The indicators V1 and V2 can be stated as
 T.C. Kuo et al. / Journal of Cleaner Production 139 (2016) 337e346 341

Table 4 Lower Higher

Carbon taxes for different levels of carbon emission. Years Years
carbon tax carbon tax
Carbon emission level Carbon emission Carbon tax (per 1000 tons) M = 800 10 20 30 40 50 M = 800 10 20 30 40 50
Level 1 <2 million tons 500 NTD 100.4 100.4

Carbon emission

Carbon emission
Level 2 2 million~4 million tons 600 NTD SASX SBSZ SASX
Level 3 4 million~6 million tons 700 NTD 112.95 112.95
SBSX
Level 4 6 million~8 million tons 800 NTD 125.5 125.5 SASZ
Level 5 8 million~10 million tons 900 NTD SBSX SASZ
Level 6 >10 million tons 1000 NTD 138.05 SA S Z & & 138.05
SASZ SBSZ
Remark: The information of carbon tax rate is available in (Kaohsiung, 2009). 150.6 150.6 SBSX & SASZ

P S Lower Higher
fCPY b $C Yf ¼ð f
a$NY Þþ1 ðCD Þ g: the cost of carbon tax incurred in the Years Years
carbon tax carbon tax
incomplete year interval during the period of production
quantity reduction M = 60,000 10 20 30 40 50 M=60,000 10 20 30 40 50
fCPY b $C$g$ðCD ÞYS g: the cost of carbon tax incurred in complete 100.4 100.4
SASX

Carbon emission

Carbon emission
SBSZ
year intervals after the period of production quantity reduction 112.95 112.95
P
V2 di¼bþ1 fCPYi $C$NY $ðCD ÞYS g: the cost of carbon tax incurred in
125.5 SASZ & SBSZ 125.5 SASZ
complete year intervals after the period of production quantity
reduction 138.05 138.05 SASZ SBSX &
& SBSZ
V3 fCPYε $CðS  d$NY ÞðCD ÞYS g: the cost of carbon tax incurred in 150.6 150.6 SASZ
the incomplete year interval after the period of production
quantity reduction. Unit
M: Thousand NTD
Carbon emission: Thousand Ton
g ¼ minfS  YS ; b$NY  YS g:
Fig. 1. The adopted equilibrium pairs in the electronics industry.
To identify the number of complete year intervals, we have

Ys V2 ¼ 0 and V3 ¼ 0 ; if q  b ¼ 0
a ¼ b ¼ Y; g ¼ S  Ys ; and d ¼ ε ¼ 0 if Y V2 ¼ 0 and V3 ¼ 1 ; if q  b ¼ 1
NY
V2 ¼ 1 and V3 ¼ 1 ; if q  b  2:
 
Ys Ys
a¼ and b ¼ a þ 1 ; if Ys mod NY s0 and <Y
NY NY

  2.2.3. SB versus SX
Y Ys In the pair of SB and SX, the enterprise reduces the carbon
a¼ s and b ¼ a ; if Ys mod NY ¼ 0 and < Y:
NY NY emission by improving its production process flow or replacing
production equipment (SX), and the government adopts the policy
Additionally,
that the unit carbon tax depends on the carbon emission level (SB).
  The enterprise profit and the government payoff can be stated as
S Ys
q¼ þ 1 ; if S mod NY s0 and <Y 8
NY NY b <
    XS
t
X
S Ys Enterprise profit EPSB SX ¼ PF  M ð1 þ IÞ  C $C
q¼ ; if S mod NY ¼ 0 and < Y: : PLi
t¼1 i¼ε
NY NY 9
X1
fiεþ1 =
To compute g, d, and ε, we calculate q-b g
 1  MD $MR
;
g¼fiε
S
g ¼ b$NY  YS ; and d ¼ ε ¼ Y ; if  Y and q  bs0
NY ( )
Pb Pfiεþ1 1  g
The indicators V1, V2, and V3 can be stated as C$S i¼ε
C g¼f 1MD $MR
  iε

Governmentpayoff GTSB SX ¼ ( )
V1 ¼ 0 ; if a ¼ 0 Pb Pfiεþ1 1  g
CPLi $C g¼f 1MD $MR
V1 ¼ 1 ; if a > 0 i¼ε iε

where

Table 5
Levels of carbon taxes for different levels of carbon emission.

Carbon emission level Carbon emission Carbon tax (per thousand tons)

Electronics industry Cement industry Steel industry

Level 1 <2 million tons 40,000 NTD 10,000 NTD 14,500 NTD
Level 2 2 million~4 million tons 48,000 NTD 12,000 NTD 17,400 NTD
Level 3 4 million~6 million tons 56,000 NTD 14,000 NTD 20,300 NTD
Level 4 6 million~8 million tons 64,000 NTD 16,000 NTD 23,200 NTD
Level 5 8 million~10 million tons 72,000 NTD 18,000 NTD 26,100 NTD
Level 6 >10 million tons 80,000 NTD 20,000 NTD 29,000 NTD
342 T.C. Kuo et al. / Journal of Cleaner Production 139 (2016) 337e346

PF ¼ S$Q$P: the total sales revenue the carbon emission level (SB). The enterprise profit and the gov-
P
M St¼1 ð1 þ IÞt : the cost of improving production process flow ernment payoff can be stated as
or replacing production equipment
Pb Pfiεþ1 1
i¼ε
fCPLi $C g¼f ð1  MD $MRg Þg: the cost of carbon tax
iε

The required parameters in this case, ε, b, and fi, can be stated as

8 9
<P 8
<
n
PfðdiÞþ1 1
9 o=
d g= YS
C$S  C g¼f ðCD Þ ; þ CðS þ 1  YS ÞðCD Þ
  : i¼b : ðdiÞ ;
Government payoff GTSB SY ¼ P 8 < P
9
=
n o
d fðdiÞþ1 1
i¼b :CPLi $C g¼f
ðCD Þg ; þ CPLb $CðS þ 1  YS ÞðCD ÞYS
ðdiÞ

8 9
2  3   Xd < X1
fðdiÞþ1 =
C 1  MD $MR0 Enterprise profit EPSB SY ¼ PF  CPLi $C ðCD Þg
: ;
ε¼4 5þ1 i¼b g¼fðdiÞ
NL n o
 CPLb $CðS þ 1  YS ÞðCD ÞYS
The calculation of b that depends on ε can be stated as
where
b¼Lande¼0 ;if εL
2  3 P S
C 1MD $MRS1 PF ¼ Q $P Yg¼1 fðCD Þg ð1 þ ðð1  ðCD Þg Þ=Es ÞÞg þ Q
b¼L ;if 4 5þ1Landε<L ðCD Þ $Pð1 þ ðð1  ðCD ÞYS Þ=Es ÞðS  YS Þ: the total sales revenue
Pd
YS
PfðdiÞþ1 1
NL ðCD Þg g: the cost of carbon tax during the
i¼b fCPLi $C g¼fðdiÞ
2  3 2  3 period of production quantity reduction
C 1MD MRS1 C 1MD $MRS1 fCPLb $CðS þ 1  YS ÞðCD ÞYS g: the cost of carbon tax after the
b¼4 5þ1;if 4 5þ1<Landε<L:
period of production quantity reduction.
NL NL

The calculations of b, d, and e are

When ε< L, the level difference e between b and ε is e ¼ bε. We

CðCD ÞYS
b ¼ d ¼ L and e ¼ 0 ; if þ1L
NL
" # " # " # " #
CðCD ÞYS CðCD Þ1 CðCD Þ1 CðCD ÞYS
b¼ þ 1 and d ¼ þ 1 ; if þ 1 < L and þ 1<L
NL NL NL NL
" # " # " #
CðCD ÞYS CðCD Þ1 CðCD ÞYS
b¼ þ 1 and d ¼ L ; if þ 1  L and þ 1 < L:
NL NL NL

YS
define fi, i ¼ 0, 1, …, eþ1 as When CðCNDLÞ þ 1 < L, the level difference e between b and d is
e ¼ bd. We define fi, i ¼ 0, 1, …, eþ1 as
f0 ¼ 0;
f0 ¼ 0;

N ½ðε  1Þ þ a
fa ¼ logMR 1 L  logMR MD þ 1; a ¼ 1; 2; …e 0 1
1
C NL CðCNDL Þ  a þ 1
fa ¼ logCD @ Aþ1 ; a ¼ 1; 2; …e
C
feþ1 ¼ S:
feþ1 ¼ YS :

2.2.4. SB versus SY
In the pair of SB and SY, the enterprise reduces the carbon 2.2.5. SA versus SZ
emission by cutting down its production quantity (SY), and the In the pair of SA and SZ, the enterprise simply chooses to pay the
government adopts the policy that the unit carbon tax depends on carbon tax without attempting to reduce carbon emission (SZ), and
 T.C. Kuo et al. / Journal of Cleaner Production 139 (2016) 337e346 343

2.2.6. SB versus SZ
SASX & SBSZ In the pair of SB and SZ, the enterprise simply chooses paying the
Lower Higher carbon tax without attempting to reduce carbon emission (SZ), and
Years Years
carbon tax carbon tax the government adopts the policy that the unit carbon tax depends
M = 21,130 10 20 30 40 50 M = 21,130 10 20 30 40 50 on the carbon emission level (SB). The enterprise profit can be
2,680 2,680 stated ashere
Carbon emission

Carbon emission
SBSZ
3,015 SASX 3,015
SBSX PF ¼ S$Q$P: the total sales revenue
3,350 SASZ 3,350 SASZ SASX
& CPLa$C$S: the cost of carbon tax.
3,685 SASZ SBSX 3,685
SBSZ
&
4,020 4,020 SASZ The level of carbon emission throughout the planning horizon is
SASZ
SASX & SBSY SBSX & SASZ C C
Lower Higher a¼ þ1 ; if þ1L
Years Years NL NL
carbon tax carbon tax
C
M = 422,600 10 20 30 40 50 M=422,600 10 20 30 40 50 a ¼ L ; if þ 1 > L:
NL
2,680 2,680
Carbon emission

Carbon emission

SBSZ Similar to Section 2.3.5, the payoff of the government is trivially

3,015 3,015 SASX
set to zero in Scenario Z (SZ) due to zero reduction in the carbon
3,350 SASZ & SBSZ 3,350 SASZ
& emission.
3,685 3,685 SASZ SASZ
S S
& B Z
4,020 4,020 SBSX 3. Case study and analysis

Unit 3.1. Overview and input data

M: Thousand NTD
Carbon emission: Thousand Ton
This section describes a case study of the strategic responses of
Fig. 2. The adopted equilibrium pairs in the cement industry. three firms in Taiwan's electronics, cement, and steel industries to
two carbon emission policies promulgated by government. The
three industries are characterized by the highest levels of pollution
the government adopts the policy that the unit carbon tax increases
and energy consumption during production (CEPD, 2008; Chou and
over time in the planning horizon (SA). The enterprise profit can be
Liou, 2012). All currency is denominated in New Taiwan Dollars
stated as
(NTD). The posted bank interest rate (I) in Taiwan Central Bank is
X 1.68% (BOT, 2012). The depreciation rate (MR) in Scenario X is
  a
Enterprise Profit EPSA SZ ¼ PF  V1 fCPYi $C$NY g assumed to be 3% (GQM, 2012) and the goal of 25% reduction in the
i¼1 carbon emission can be achieved via Scenario X. In Scenario Y, the
n o
annual carbon reduction ratio, (CD), can be approximated as 9%,
 CPY b $CðS  a$NY Þ
since the goal of 25% (1-((1e0.09)3)100%) carbon reduction needs
to be achieved over three years (AUO, 2012), i.e. the production
where

PF ¼ S$Q$P: the total sales revenue Lower Higher

P Years Years
V1 ai¼1 fCPYi $C$NY g: the cost of carbon tax incurred in complete carbon tax carbon tax
year intervals M = 133,830 10 20 30 40 50 M = 133,830 10 20 30 40 50
fCPY b $CðS  a$NY Þg: the cost of carbon tax for the remaining in
16,976 16,976
Carbon emission

Carbon emission

an incomplete year interval. SASX

19,098 19,098
SBSX
To identify a and b, we have 21,220 21,220 SASZ SASX
23,342 23,342
S S SASZ
a¼ ; and b ¼ a þ 1 ; if þ1Y 25,464 25,464
NY NY
S SBSX & SASZ
a ¼ b ¼ Y ; if þ 1 > Y: Lower Higher
NY Years Years
carbon tax carbon tax
The indicators V1 can be stated as M =5,353,200 10 20 30 40 50 M=5,353,200 10 20 30 40 50
16,976 16,976 SBSZ
V1 ¼ 0 ; if a ¼ 0
Carbon emission

Carbon emission

SASX
V1 ¼ 1 ; if a > 0: 19,098 19,098
SBSX
The payoff of the government is the decrease in the total carbon 21,220 SASZ & SBSZ 21,220

emission due to a unit of carbon taxes collected by the government. 23,342 23,342 SSBBSX
In Scenario Z (SZ), the enterprise does not make any attempts to SASZ &
&
25,464 25,464 SSAASSZZ
reduce the carbon emission. The metric to evaluate the government
payoff is the decrease in the total carbon emission due to a unit of Unit
M: Thousand NTD
carbon taxes collected by the government. Therefore, the payoff of Carbon emission: Thousand Ton
the government is trivially set to zero in Scenario Z (SZ) due to zero
reduction in the carbon emission. Fig. 3. The adopted equilibrium pairs in the steel industry.
344 T.C. Kuo et al. / Journal of Cleaner Production 139 (2016) 337e346

quantity is reduced by 9% in the first, second, and third years in the emission, whereas Scenario Z implies that the enterprise only pays
planning horizon to achieve the goal of 25% carbon reduction. The carbon taxes for compensation of carbon emission rather than
elasticity measures in the electronics, cement, and steel industries taking any action to reduce the carbon emission. Figs. 1e3 reveal
are 0.950 (WLCS, 2006), 0.906 (Tcement, 2006), and 0.998 (CSC, several important managerial insights for policy-makers.
2006), respectively.
In the electronics industry, the product type is television sets. 3.3. Managerial insights
The average selling price per set is 21,888 NTD (Sesame blog, 2009),
the annual sales quantity is 100,000 (UDN News Net, 2009), and the We gain four insights from the results of our case study as
annual carbon emission is 125,500 tons (TPCF, 2012). In the cement follows.
industry, the product type is cement. The average selling price of
cement per ton is 2050 NTD (MoneyDJ, 2010), the annual sales (i) In the situation of the low carbon tax (CPYi and CPLi) and low
quantity is 10 million tons (YNWC, 2010), and the annual carbon expense (M) for improving production processes, the gov-
emission is 3.35 million tons (YNWC, 2010). In the steel industry, ernment most likely chooses Scenario A. However, in the
the product type is steel. The average selling price of steel per ton is same situation, an unfavorable outcome results when an
24,160 NTD (SP, 2012), the annual sales quantity is 9.73 million tons enterprise chooses not to reduce carbon emission. This
(CT, 2012), and the annual carbon emission is 21.22 million tons outcome should alert policy-makers that even under a low
(YNWC, 2010). capital investment in improving production processes to
Following Chiang (2012), we set 10 years as a year interval in reduce carbon emission, the enterprise may still end up with
Scenario A, i.e. NY ¼ 10. The carbon tax (CPYi, i ¼ 1, 2, 3) is suggested Scenario Z.
as 200/500/1000 NTD per thousand ton of carbon emission for the (ii) In the situation of the low carbon tax (CPYi and CPLi) and high
first-, second-, and third-year intervals, i.e. 200, 500, and 1000 NTD expense (M) for improving production processes or replacing
per thousand ton of carbon emission are collected by the govern- production equipment, the enterprise, as expected, chooses
ment in the 1e10, 11e20, and 21e30 years. In Scenario B, six levels Scenario Z, no matter which of the two policies the govern-
(L) of the quantity of carbon emission are considered and the gap of ment adopts. This situation demonstrates that the carbon tax
each level (NL) is set as 2 million tons (Kaohsiung, 2009). The carbon mechanism may not always result in reducing emission as
tax (CPLi, i ¼ 1, …, 6) per thousand ton for the first level is 500 NTD envisioned by policy-makers.
and its increment per thousand tons is 100 NTD (Kaohsiung, 2009). (iii) Compared to the situation of the low carbon tax (CPYi and
Table 4 lists the taxes for the different levels of carbon emission. CPLi) and low expense (M), the government prefers Scenario B
to Scenario A (high carbon tax and high expense of
3.2. Case study results improving production processes). Meanwhile, the enterprise
seems to choose Scenario X (capital investment to improve
We study how the expense for modifying production processes production processes or replacing old production equip-
or equipment (M) and carbon tax (CPYi and CPLi) affect the equilib- ment), or Scenario Z (no action). The case study finds only a
rium pair of the scenarios adopted by the government and enter- few occasions of Scenario Y (reduction in the production
prises in the three industries. Due to data unavailability of M, we quantity).
conduct a sensitivity analysis of M. The low and high levels of (iv) When enterprises choose to pay taxes rather than reducing
expense of improving the production process or replacing pro- emission, this unfavorable outcome may “force” the gov-
duction equipment in the electronics, cement, and steel industries ernment to offer other incentives. The case study shows that
are 800 and 60,000 thousand NTD, 21,130 and 422,600 thousand enterprises tend not to choose Scenario Y (reduction in the
NTD, and 133,830 and 5,353,200 thousand NTD, respectively. The production quantity), because the reduction in production
carbon tax data presented in Section 3.1 are the low level and we quantity may in fact erode profitability. However, if the
set the high level of carbon taxes (CPYi, i ¼ 1, 2, 3) as 14,000/35,000/ reduction in carbon emission can be achieved under an
70,000 NTD per thousand tons of carbon emission (electronics in- affordable capital investment on M and the carbon tax level
dustry), 4000/10,000/20,000 NTD per thousand tons of carbon is high enough, it is likely that enterprises will improve their
emission (cement industry), and 10,200/25,500/51,000 NTD per production processes or replace old production equipment.
thousand tons of carbon emission (steel industry) for the first-,
second-, and third-year intervals in the three industries. Table 5 4. Conclusions
lists the high levels of carbon tax (CPLi, i ¼ 1, …, 6) per thousand
tons for the different levels of carbon emission. A carbon tax policy is considered an effective mechanism for
Figs. 1e3 show the equilibrium pair of the policy adopted by the reducing global carbon emission, but appropriate levels of tax may
government and the strategy chosen by the enterprise for the three in fact induce carbon-producing enterprises to change their pro-
industries. The first notation in the pair S.S. is the policy adopted by duction process for emission reduction in their long-term business
the government and the second notation is the strategy chosen by strategies. While prior studies focus on the national level to explore
the enterprise, e.g., SASX denotes the equilibrium pair of the gov- the impacts of carbon tax on the behaviors of firms and customers,
ernment policy and the strategy of the enterprise as Scenario A and this study discusses the effects of various carbon tax policies on
Scenario X, respectively. In each sub-figure, we study how the three enterprises in the high-carbon, high-energy electronics,
different planning horizon of implementation of Scenarios A and B cement, and steel industries in Taiwan. The investigated results
and the different total carbon emission vary the results of the show that the enterprises are most likely to modify their produc-
adopted equilibrium pairs. tion processes for carbon emission if the government applies the
In this case study, neither of the government's two policies correct tax.
prevails. However, the government may prefer Scenarios X to Y, We view the government and the enterprises as independently
followed by Scenario Z chosen by the enterprises, since the gov- seeking their own optimums. We analyze the Nash equilibrium pair
ernment wants them to improve their production processes or of the policy adopted by the government and the strategy chosen by
replace their production equipment. In Scenario Y, the enterprises the enterprise, where no player can be better off by a unilateral
choose to reduce the production quantity to reduce the carbon change in its decision. We discuss how two carbon tax policies
 T.C. Kuo et al. / Journal of Cleaner Production 139 (2016) 337e346 345

adopted by the government affect the three strategic responses by CSC (China Steel Corporation annual report), 2006, http://www.csc.com.tw/csc/ss/
year.htm, (accessed on 15.07.12.) (in Chinese).
the enterprises, i.e. investing in new technology (the improvement
CSGCC (The Chinese Study Group of Climate Change), 2000. Chinese Study of
of production processes or replacement of old production equip- Climate Change. Tsinghua Publication, Beijing.
ment), reducing production quantity to decrease carbon emission, CT (China Times), 2012, http://tw.news.yahoo.com/%E5%B0%8F%E8%B3%BA3-29%
and remitting the tax rather than reducing emission. A case study E5%84%84-%E4%B8%AD%E9%8B%BC3%E6%9C%88%E8%99%A7%E8%BD%89%E7%
9B%88-213000478.html, (accessed on 13.07.12.).
numerically demonstrates that an inappropriate carbon tax level Datta, A., 2010. The incidence of fuel taxation in India. Energy Econ. 32 (Suppl. 1),
results in an unintended consequence, i.e. that enterprises may S26eS33.
choose to pay carbon taxes for compensation of carbon emission Drouet, L., Labriet, M., Loulou, R., Vielle, M., 2008. A Master Program that Will Drive
the Coupling of GEMINI-E3 and MARKAL TIMES Models (Working paper).
rather than attempting to reduce carbon emission. As expected, in Elliott, J., Fullerton, D., 2014. Can a unilateral carbon tax reduce emissions else-
the situation of the low carbon tax and high expense for improving where? Resour. Energy Econ. 36, 6e21.
production processes or replacing old production equipment, no EPA, 2013. EPA Fact Sheet: Reducing Carbon Pollution from Power Plants. http://
www2.epa.gov/sites/production/files/2013-09/documents/20130920factsheet.
matter what scenarios the government adopts, the enterprises do pdf (accessed on 02.05.14.).
not choose to reduce their carbon emission, due to the likelihood of Fahimnia, B., Sarkis, J., Choudhary, A., Eshragh, A., 2015. Tactical supply chain
high investment costs and relatively low carbon taxes. Given the planning under a carbon tax policy scheme: a case study. Int. J. Prod. Econ. 164,
206e215.
low expense for improving production processes or replacing old García Benavente, J.M., 2016. Impact of a carbon tax on the Chilean economy: a
production equipment, our case study still shows that an enterprise computable general equilibrium analysis. Energy Econ. 57, 106e127.
will not reduce carbon emission even if the government collects the Good Quality Machinery (GQM), 2012, http://www.goodqualitymachinery.com.tw/
energy-02.html, (accessed on 16.07.12.) (in Chinese).
low level of tax. We conclude, therefore, that policy-makers may
Guo, Z., Zhang, X., Zheng, Y., Rao, R., 2014. Exploring the impacts of a carbon tax on
need to provide other incentives for compensation of the expense the Chinese economy using a CGE model with a detailed disaggregation of
for improving production processes, or impose a considerable tax energy sectors. Energy Econ. 45, 455e462.
penalty if the enterprise decides not to curtail its carbon emission. Gupta, M., 2016. Willingness to pay for carbon tax: a study of Indian road passenger
transport. Transp. Policy 45, 46e54.
Having focused on firms' process flow improvement and pro- Huisingh, D., Zhang, Z., Moore, J.C., Qiao, Q., Li, Q., 2015. Recent advances in carbon
duction curtailment strategies in response to government-imposed emissions reduction: policies, technologies, monitoring, assessment and
carbon taxes policies, we suggest that future research should modeling. J. Clean. Prod. 103, 1e12.
Hua, G., Cheng, T.C.E., Wang, S., 2011. Managing carbon footprints in inventory
consider other strategies including changing production supply management. Int. J. Prod. Econ. 132, 178e185.
chains, and transferring suppliers to countries with lower carbon Hwang, J.J., 2011. Policy review of greenhouse gas emission reduction in Taiwan.
taxes. Furthermore, carbon accounting to assist management in Renew. Sustain. Energy Rev. 15, 1392e1402.
Johansson, B., 2000. Economic Instruments in Practice 1: Carbon Tax in Sweden.
setting appropriate emission reduction targets should be integrated http://www.oecd.org/dataoecd/25/0/2108273.pdf (accessed on 19.07.12.).
with the research results discussed in this paper. Kaohsiung, 2009. The Hearing Minute of Kaohsiung City Government for Carbon
Tax, pp. 231e235 (in Chinese).
Kim, H.G., 2008. On the views of carbon tax in korea. Am. J. Appl. Sci. 5 (11),
Acknowledgments 1558e1561.
Lee, C.F., Lin, S.J., Lewis, C., Chang, Y.F., 2007. Effects of carbon taxes on different
industries by fuzzy goal programming: a case study of the petrochemical-
The authors thank the reviewers and the editor-in-chief for their related industries, Taiwan. Energy Policy 35, 4051e4058.
helpful and valuable comments and are grateful for the generous Lee, C.F., Lin, S.J., Lewis, C., 2008. Analysis of the impacts of combining carbon
interaction and guidance provided by many industry experts. We taxation and emission trading on different industry sectors. Energy Policy 36
(2), 722e729.
are also grateful for the editorial services of Ann Stewart. The au-
Lin, B., Xie, X., 2016. CO2 emissions of China's food industry: an input-output
thors thank the College of Electrical Engineering and Computer approach. J. Clean. Prod. 112, 1410e1421.
Science at Chung Yung Christian University for partially supporting Liu, Z.-J., Sun, D.-P., Lin, C.-X., Zhao, X.-Q., Yang, Y., 2016. Multi-objective optimi-
this research under grant CYCUEECS-10001, and the Ministry of zation of the operating conditions in a cutting process based on low carbon
emission costs. J. Clean. Prod. 124, 266e275.
Science and Technology, Taiwan for financially supporting this Lu, C., Tong, Q., Liu, X., 2010. The impacts of carbon tax and complementary policies
research under grants NSC99-2621-M-033-002-MY3, NSC102- on Chinese economy. Energy Policy 38 (11), 7278e7285.
2221-E-002-108-MY3, NSC102-2221-E-002-108-MY3, and MOST Ma, X., Ji, P., Ho, W., Yang, C.-H., 16 March, 2016. Optimal procurement decision with
a carbon tax for the manufacturing industry. Comput. Oper. Res.
103-2221-E-002-217-MY2. The second author thanks National Meng, S., Siriwardana, M., McNeill, J., 2013. The environmental and economic
Taiwan University for partial financial support under grant impact of the carbon tax in Australia. Environ. Resour. Econ. 54, 313e332.
104R7768. Both authors are grateful to Chien Ying Lee, Ming Yu MoneyDJ, 2010, http://www.moneydj.com/kmdj/news/newsviewer.aspx?
a¼e85e22a8-10de-47e1-b4a9-929a6a8c457a, (accessed 11.07.12). (in Chinese).
Kuo, Miao-Pei Chen, and Zi-Hao Lin for their data collection and Nakata, T., Lamont, A., 2001. Analysis of the impacts of carbon taxes on energy
analysis. systems in Japan. Energy Policy 29, 159e166.
Ockenden, W., McAloon, C., 2011. Rural Report for Gippsland and South-east New
South Wales.
References Porter, M.E., van der Linde, C., 1995. Toward a new conception of the environment-
competitiveness relationship. J. Econ. Perspect. 9 (4), 97e118.
Allan, G., Lecca, P., McGregor, P., Swales, K., 2014. The economic and environmental Porter, M.E., 1991. America's Green Strategy, in: Scientific American, p. 168.
impact of a carbon tax for Scotland: a computable general equilibrium analysis. Ruamsuke, K., Dhakal, S., Marpaung, C.O.P., 2015. Energy and economic impacts of
Ecol. Econ. 100, 40e50. the global climate change policy on Southeast Asian countries: a general
AUO News Press (AUO), 2012, http://auo.com/?sn¼21&lang¼zh-TW&c¼1&n¼1371, equilibrium analysis. Energy 81, 446e461.
(accessed on 16.07.12.) (in Chinese). Saddler, H., Muller, F., Cuevas, C., 2006. Competitiveness and Carbon Pricing Border
Bank of Taiwan (BOT), 2012, http://rate.bot.com.tw/Pages/TWN001/TWN001.aspx, Adjustments for Greenhouse Policies. Discussion Paper, p. 86.
(accessed on 16.07.12.) (in Chinese). Schaltegger, S., Synnestvedt, T., 2002. The link between green and economic suc-
Baranzini, A., Goldembergc, J., Speckd, S., 2000. A future for carbon taxes. Ecol. Econ. cess: environmental management as the crucial trigger between environmental
32 (3), 395e412. and economic performance. J. Environ. Manag. 65 (4), 339e346.
Bruvoll, A., Larsen, B.M., 2004. Greenhouse gas emissions in Norway. Do carbon Scrimgeour, F., Oxley, L., Fatai, K., 2005. Reducing carbon emissions? the relative
taxes work? Energy Policy 32 (4), 493e505. effectiveness of different types of environmental tax: the case of New Zealand.
CEPD (Council of Economic Planning and Development), 2008, http://www.ndc.gov. Environ. Model. Softw. 20 (11), 1439e1448.
tw/att/0018977/0018977.pdf, (accessed on 07.05.14.) (in Chines). Sesame blog, 2009, http://www.calldoor.com.tw/myblog/ppw102/articles/55994,
Chaabane, A., Ramudhin, A., Paquet, M., 2012. Design of sustainable supply chains (accessed on 11.07.12.) (in Chinese).
under the emission trading scheme. Int. J. Prod. Econ. 135, 37e49. Shahbaz, M., Hye, Q.M.A., Tiwari, A.K., Leit~ ao, N.C., 2013. Economic growth, energy
Chiang, K.-C., 2012. The Low Carbon Development and Low Carbon Market. http:// consumption, financial development, international trade and CO2 emissions in
www.zgkjtz.com/show.php?contentid¼169 (accessed on 13.07.12.). Indonesia. Renew. Sustain. Energy Rev. 25, 109e121.
Chou, K.T., Liou, H.M., 2012. Analysis on energy intensive industries under Taiwan's Siriwardana, M., Meng, S., McNeill, J., 2011. The Impact of a Carbon Tax on the
climate change policy. Renew. Sustain. Energy Rev. 16, 2631e2642. Australian Economy: Results from a CGE Model. Business, Economics and Public
346 T.C. Kuo et al. / Journal of Cleaner Production 139 (2016) 337e346

Policy Working Papers. net/ODM-.htm, (accessed on 16.07.12.) (in Chinese).

SP (Steel Price), 2012, http://jsjustweb.jihsun.com.tw/z/ze/zeq/zeq.asp.htm, Young, C.Y., Huang, W.M., 2012. Review of Taiwan's climate policy after Copenha-
(accessed on 16.07.12.) (in Chinese). gen. Renew. Sustain. Energy Rev. 16, 20e28.
Tcement (Taiwan Cement Annual report), 2006, http://www.taiwancement.com/ YNWC (Youth: the Next Wave for Change), 2010, http://www.earthday.org.tw/
#Invest_4_1_1, (accessed on 16.07.12.) (in Chinese). node/2847, (accessed on 11.07.12.) (in Chinese).
TPCF (Taiwan Product Carbon Footprint), 2012, http://cfp.epa.gov.tw/carbon/ Zhang, Z.X., Baranzinic, A., 2004. What do we know about carbon taxes? an inquiry
ezCFM/Function/PlatformInfo/FLabelProduct/FLProductInfo.aspx, (accessed on into their impacts on competitiveness and distribution of income. Energy Policy
11.07.12.) (in Chinese). 32 (4), 507e518.
UDN News Net, 2009, http://pro.udnjob.com/mag2/tool/storypage.jsp?f_ART_ Zimmer, A., Jakob, M., Steckel, J.C., 2015. What motivates Vietnam to strive for a
ID¼51092, (accessed on 11.07.12.) (in Chinese). low-carbon economy? d On the drivers of climate policy in a developing
World-wide laptop computers sales (WLCS), 2006, http://legend34.myweb.hinet. country. Energy Sustain. Dev. 24, 19e32.