Subscriber access provided by READING UNIV
Article
Multi-regional Input-output Analysis of Spatial-temporal Evolution Driving Force for Carbon Emissions Embodied in Interprovincial Trade and Optimization Policies: A Case Study of Northeast Industrial District in China Hao Cheng, Suocheng Dong, Fujia Li, Yang Yang, Shantong Li, and Yu Li Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b04608 • Publication Date (Web): 05 Dec 2017 Downloaded from http://pubs.acs.org on December 10, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 33
Environmental Science & Technology
1
Multi-regional Input-output Analysis of Spatial-temporal
2
Evolution Driving Force for Carbon Emissions Embodied in
3
Interprovincial Trade and Optimization Policies: A Case Study
4
of Northeast Industrial District in China
5 6
Hao Cheng,a Suocheng Dong,a Fujia Li,a,* Yang Yang,a,b Shantong Li,c and Yu Lia
7
(a. Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China; b. University of Chinese Academy of
8
Sciences, Beijing 100049, China; c. Department of Development Strategy and Regional Economy, Development Research Center, State
9
Council, Beijing 100010, China)
10 11
ABSTRACT: In the counties with rapid economy and carbon emissions(CEs) growth, CEs embodied in
12
interprovincial trade(CEs-PT) significantly impacts the CEs amount and structure, and represents a key issue to
13
consider in CEs reduction policies formulation. This study applied EEBT and two-stage SDA model to analyze the
14
characteristics and driving force of spatial-temporal evolution for net CEs-PT outflow in the Northeast Industrial
15
District of China(NID). We found that, during 1997-2007, the net CEs-PT flowed out from NID to 16 south and
16
east provinces, then to 23 provinces all over China, and its amount has increased 216.798Mt(by 211.67% per year).
17
The main driving forces are technology and demand(further decomposed into structure and scale matrix), the
18
contribution are 71.6418Mt and 145.1562Mt. Then, we constructed coupling relationship model and took the top
19
three industries with the greatest net CEs-PT outflow(Farming, forestry, animal husbandry and fisheries,
20
Electricity and heat production and supply, and Petroleum processing, coking and nuclear fuel processing) as
21
examples, adjusted the interprovincial trade constructions, scales and objects, to reduce the CEs-PT with lower
22
costs, greater effect and more equitable. The achievement could provide reference for formulating CEs reduction
23
policies for similar areas in the world characterized by rapid growth of economy and CEs.
24 25
TOC Art
ACS Paragon Plus Environment
Environmental Science & Technology
Page 2 of 33
26
1. INTRODUCTION
27
Carbon emissions embodied in interregional trade (CEs-RT) has a significant impact on the
28
amount and structure of regional carbon emissions (CEs), and it is an important factor that should
29
be considered in any regional CEs reduction policies formulation and adjustment. Currently, this
30
issue has attracted much attention in the literature,
31
CEs-RT have proliferated.
32
international trade (CEs-NT), while the CEs embodied in interprovincial trade (CEs-PT) within a
33
single country has received not enough attention.
1
2
3,4
5,6
and studies focusing on the transference of
However, most studies have analyzed the CEs embodied in
7-10
34
In fact, for countries with larger economies and greater CEs, they usually have more
35
provinces and more active interprovincial trade; and the CEs-PT has a greater influence on the
36
CEs in an administrative region than the CEs-NT,
37
than between countries. To engage with this issue, some scholars have conducted beneficial
38
explorations of the CEs-PT in different countries. McGregor et al. analyzed the CO2 pollution
39
content of interregional trade flowed between Scotland and the rest of the UK and found that the
40
interregional environmental spillovers within the UK was significant and that a CO2 ‘trade balance’
41
existed between Scotland and the rest of the UK.
42
model to track interregional carbon flowed in the Jing−Jin−Ji Area by combining network analysis
43
and input-output analysis, and found that the CO2 emissions embodied in products was only partially
44
controlled by a region and that the controlled carbon accounted for approximately 70% of the total
45
embodied carbon.
46
relationship inter-regional spillover of CO2 emissions and domestic supply chains for 2002 and
47
2007 within eight regions of China.
48
obtain a regional map of carbon footprints within eight regions of China from 1997 to 2007.
49
Zhang et al. applied EEBT model to clarified the provincial and sectoral contributions to national
50
emissions. Zhang et al. used provincial-level MRIO model to study the trends and disparities of
51
consumption-based emissions from Chinese provinces during the period of 2002-2007. Liu et al.
52
provided a dynamic analysis of carbon emissions embodied in consumption and export demand-
13
11
12
as trade barriers are lower between provinces
Chen and Chen presented a hybrid network
Meng et al. applied the inter-regional input-output model to explain the
14
Tian et al. applied carbon footprints and SDA analysis to 15
16
17
ACS Paragon Plus Environment
Page 3 of 33
Environmental Science & Technology
11
53
supply chains at the sub-national level based on the MRIO tables for 1997 and 2007. Zhang and
54
Tang used MRIO model and logarithm mean Divisia index approach to analyze the changes in
55
China's carbon embodied in exports at the national and provincial levels. It is noteworthy that Su
56
and Ang have conducted a continuing study of the CEs-RT, at the level of spatial aggregation and
57
sector aggregation, from the national scale and the global scale.
58
to be meaningful for revealing CEs-PT transference.
59
18
19-24
These studies are considered
The transfer quantity and direction have attracted the bulk of the attentions in the CEs-PT 25-28
60
researches
61
transference, while little research has implemented optimization policies based on the driving
62
force mechanisms analysis. What’s more, for the lack of such studies, the researches on
63
formulating policies to reduce CEs effect have been limited. Both the depth and the amount of the
64
studies are far from meeting the actual needs of CEs reduction. As a result, in most countries,
65
particularly those with greater CEs, the formulation of CEs reduction policies lacked provincial
66
coordination and cooperation, and the reduction effect was much weaker for the interference and
67
restriction of the CEs-PT. These policies reduced the CEs of a region or an industry on the surface,
68
while actually increased the CEs in another region or industry through trade associations.
69
these policies reduced the CEs reduction effect or even generated a negative effect, wasted CEs
70
reduction investment and created an unfair distribution of reduction responsibility.
71
gap in the literature, we urgently need to analyze the spatial-temporal evolution of the CEs-PT
72
transference over a long period, explore the sources and driving force mechanisms, and implement
73
optimization policies to coordinate the relationship between interprovincial trade and the CEs-PT.
74
Ultimately, the result could guide policymaker to formulate CEs reduction policies with lower
75
costs, better reduction effects, more equitable and greater cooperation between regions.
, but these researches have less consideration for the driving force of the CEs-PT
30,31
29
Thus,
To fill this
76
The Northeast Industrial District of China was selected as a case study for the CEs-PT
77
research from 1997 to 2007, because in this area and during this period, the CEs had a greater
78
contribution rate of China, and GDP and CEs had grown rapidly. Based on the 1997, 2002, and
79
2007 input-output (IO) tables of 30 provinces in China, this study applied emissions embodied in
80
bilateral trade model(EEBT), to analyze the characteristics of spatial-temporal evolution for the
ACS Paragon Plus Environment
Environmental Science & Technology
81
CEs-PT transference between the study area and other provinces. Then we applied two-stage
82
structural decomposition analysis(SDA) approach to analyze the driving force for CEs-PT
83
transference, including technology driving force and demand driving force (which was
84
decomposed into structure driving force and scale driving force). Focused on these driving forces,
85
we built a coupling relationship model(CR) of net commodity value outflow change and net CEs-
86
PT outflow change to adjusted the regional industrial structure and the interprovincial trade
87
structure in order to scientifically and efficiently reduce the CEs-PT, achieve the CEs reduction
88
target with lower costs, better effects and more equitable. This study will provide a reference for
89
formulating CEs reduction policies with lower costs, better effects and more equitable in China
90
and in similar areas worldwide characterized by rapid growth in both the economy and the CEs.
91 92
2. MATERIALS AND METHODS
93
2.1. Case Study. The study area is the Northeast Industrial District of China (NID), including
94
three traditional industrial provinces of Liaoning Province, Jilin Province and Heilongjiang
95
Province in the northeast of China. The total area of the NID is 787.3 thousand square kilometers,
96
accounting for 8.2% of China. From 1997 to 2007, the GDP of the NID increased from 776.684
97
billion Yuan to 2,355.299 billion Yuan, with an average annual growth rate of 20.32%, accounting
98
for 8.86% of China’s GDP in 2007 (Figure 1). The NID had long maintained close interprovincial
99
trade relationships with other provinces in China and had made outstanding contributions to
100
China’s economy. The NID was once an important engine for the rapid economic growth in China.
101
Meanwhile, the NID had also greatly contributed to the rapid CEs growth in China. From
102
1997 to 2007, the total CEs of the NID increased from 103.0425 Mt to 364.0169 Mt, with an
103
average annual growth rate of 25.33%, accounting for 20.20% of China’s CEs in 2007 (Figure 1),
104
which far exceeded the proportions of the area and GDP. Thus, clearly the industrial structure
105
tended toward heavy industry, and the CEs reduction pressure was huge. Currently, the NID is
106
facing the double pressures of the economic downturn, as China's economic growth has slowed,
107
and the CEs reduction. It is thus urgent to formulate scientific and reasonable CEs reduction
108
policies to improve the reduction effect with the reduction cost in minimize.
ACS Paragon Plus Environment
Page 4 of 33
Page 5 of 33
Environmental Science & Technology
109 110
Figure 1. GDP and CEs of the Northeast Industrial District from 1997 to 2007
111 112
Therefore, considering the characteristics of heavy industrial structure, high CEs and high
113
interregional trade volume, the NID is very typical and representative of those regions with rapid
114
growth of economy and CEs in the world. The NID was selected as a case study for assessing the
115
dynamic impact of the CEs-PT on CEs, implementing optimization policies to reduce CEs and
116
providing a reference for similar regions worldwide.
117
2.2. Data Sources. In this study, we used the 1997 IO tables (with 40 sectors for 30 provinces, 32
118
excluding Tibet)
119
Tibet)
120
Council, P.R.C.. The IO table is compiled once every five years, and the latest data was published
121
in 2007. To unify the data, we merged industries into 28 sectors.
33,34
and the 2002 and 2007 IO tables (with 42 sectors for 30 provinces, excluding
for China, which are compiled by the Development Research Center of the State
122
Owing to the lack of official CO2 emissions data at the provincial level, we estimated the
123
coefficients of CO2 emissions and energy consumption for 30 provincial regions based on the
124
IPCC reference approach.
125
different energy types used in different sectors of different regions into standard coal, in order to
126
achieve internal integration of energy type differences.
35,36
In the process of collecting energy consumption data, we converted
127
Firstly, we collected energy consumption data from the table named “Energy Consumption
128
by Sector” from 30 Provincial Statistical Yearbooks in 1997, 2002 and 2007. If there is no this
129
table, we replaced it with the table named “Overall Energy Balance Sheet (OEBS)” from
130
Provincial Statistical Yearbooks or the table named “Energy Balance of Region(EBR)” from
ACS Paragon Plus Environment
Environmental Science & Technology
Page 6 of 33
131
China Energy Statistical Yearbooks. The ECS could provide the energy consumption date of every
132
sector directly. While the OEBS or EBR could provide the energy consumption data of several
133
sectors or the total energy consumption data of some sectors. For the data given directly, it can be
134
used directly. For the total data, we decomposed it by the proportion obtained from the table
135
named “Consumption of Energy by Sector and Major Variety” or “Main Energy Consumption of
136
Industrial Enterprises above Designated Size by Sector” from Provincial Statistical Yearbooks.
137
Then we converted the actual amount of energy consumption into the standard coal, in
138
accordance with the energy conversion coefficient provided by the Intergovernmental Panel on
139
climate change (IPCC) and Energy Statistics Knowledge Manual (ESKM).
140 141
37
Finally, we converted the energy consumption data into carbon emissions data, in accordance with the carbon emissions coefficient provided by IPCC.
142
2.3. Methodology.
143
2.3.1 Multi-regional input-output analysis modeling for spatial-temporal evolution—EEBT.
144
There are two common approaches can be applied to measure embodied emissions: one
145
considers total bilateral trade between regions (EEBT approach) and the other considers trade to
146
final consumption and endogenously determines trade to intermediate consumption (MRIO
147
approach).38,39 They are both constructed based on multi-regional input-output tables(MRIO
148
tables), but they have different advantages. The EEBT model is relevant for considering the
149
environmental impacts of aggregated exports from and imports to a region. It has the
150
transparency property and is considered superior when analyzing bilateral trade and climate
151
policy. The MRIO model has the advantage of reflecting interregional spillover and feedback
152
effects. It is more applicable to the analysis of final consumption and analogous to LCA which
153
consider the total emissions from raw-material extraction to final consumption.39-41 In this
154
research, we want to formulate the optimization policies based on the interregional bilateral trade
155
and CEs-PT transference. It is too complex and unnecessary for considering the interregional
156
spillover and feedback effects. Comparing the above different advantages of two models, we
157
chose EEBT model to analyze spatial-temporal evolution of CEs-PT, for it is more suitable for
158
our research.
159
MRIO tables and their applications have generated substantial interest at the forefront of
ACS Paragon Plus Environment
Page 7 of 33
Environmental Science & Technology
160
environmental policy debate.42,43 To perform a multi-regional input-output study requires a
161
considerable amount of data, and much of these are not directly available. Therefore, various
162
approximations and simplifications should be used in the process of multi-regional input-output
163
compilation. Generally, a gravity model would be used to estimate an interregional commodity
164
flow matrix based on regional IO tables, and then the well-known iterative procedure of bi-
165
proportional adjustment of the RAS technique would be used to find a balanced multi-regional
166
input-output table.44,45 We obtained MRIO tables by analyzing interprovincial trade using the
167
cross-entropy method and a gravity model. Table 1 demonstrates the structure of the MRIO table
168
in region r. The region r is one of m regions in a country, and it has n industry sectors. In our
169
research, one province has one MRIO table in one year. So, there are total 90 MRIO tables for 30
170
provinces in China in 1997, 2002 and 2007. Based on these tables, we applied the EEBT model to
171
analyze the CEs-PT transference.
172
Table 1. Multi-regional input-output table of region r in a country with m regions(R) and n sectors(S)
Region r
Industry
Intermediate use
sector
Final
Total
Value flow out from Region r in
use
output
interprovincial trade
S1 ... Sj ... Sn Intermediate input
S1
. .. Si
. .. Sn
… … . ..
. ..
...
… … ...
. ..
. ..
… …
. ..
. ..
. ..
. ..
Added value
… …
Total input
… …
Value flow into Region R1
… …
R1
...
Rd
...
Rm
…
…
...
...
…
…
...
...
…
…
...
...
r in interprovincial
. .. Rd
trade
. .. Rm
...
. ..
...
… …
...
. ..
...
… …
173
174
where refers to intermediate input from sector i to sector j in region r (i,j=1……n, n=28 in
175
our study), is the final use of sector i, is the total output of sector i, is the added value of
ACS Paragon Plus Environment
Environmental Science & Technology
Page 8 of 33
176
sector j, and is the total input of sector j.
represents the commodity value transference of
177
sector i from region r to region d, and represents the value transference of sector j from region
178
d to region r (d,r=1……m, m=30 in our study).
179
Direct carbon emissions (DCEs) accounting, which is used under the Kyoto Protocol, is a
180
straightforward approach to accounting for CEs.44 In this study, we represented the DCEs of the
181
country as C, and the DCEs of region r as . The DCEs of region r, when divided by the total
182
output in region r, could indicate the quantity of DCEs per unit of output, namely, the direct
183
carbon emissions factor of region r, expresses as . It is calculated by the following formula:
184
= /
185
From the horizontal angles in Table 1, we obtained
186
= −
187
where X is the total output matrix, Y is the final use matrix for all regions, and A expresses
188
the local inter-industry requirements (the local technical coefficient matrix). The matrix I −
189
is generally called the Leontief inverse matrix or the cumulative coefficient matrix.
(1)
(2)
190
We used Ex instead of Y and obtained the following:
191
= − !"
192
where Ex expresses the value outflow in interregional trade, and Xe is the total input for Ex.
193
By multiplying both sides of Formula (3) by F, the CEs-PT outflow can be calculated as
194 195
(3)
follows:
= # × = # × % − !" & = ' × × % − !" &
(4)
196
where ( expresses the CEs-PT outflow, ' is the CEs coefficient, which expresses the
197
conversion from standard coal to carbon emission, and e is the energy consumption coefficient,
198
which expresses the conversion from different type energy consumption to standard coal. The e is
199
different for different fuel types.
200
The NID includes Liaoning Province, Jilin Province and Heilongjiang Province. These three
201
provinces have the similar resource endowments, economic levels, industrial structures and levels
202
of CEs in the whole trade chain. This study combined the three provinces into a composite to
203
analyze the CEs-PT transference from the NID to other provinces.
ACS Paragon Plus Environment
Page 9 of 33
Environmental Science & Technology
204 205
The CEs-PT outflow from the NID to 27 other provinces can be demonstrated by the following formula: +,
+,
+,
+,
206
)* = -* + /* + 0*
207
where 1 is the CEs-PT outflow from sector i of the NID to sector j of the 27 other
208
provinces (d=1……27, i,j=1……28), 2 expresses Liaoning Province, 3 expresses Jilin
209
Province, and 4 expresses Heilongjiang Province.
(5)
210
According to the data and the methodology described above, we could obtain the net CEs-PT
211
outflow and the net commodity value outflow in interprovincial trade of 28 sectors between NID
212
with other 27. Then net CEs-PT outflow can be calculated as follows: +,
+,
,+
213
5)* = )* − *)
214
where 67 expresses the net CEs-PT outflow from the NID to 27 other provinces, 7
215
(6)
expresses the CEs-PT outflow from the NID, and 7 expresses the CEs-PT inflow to the NID.
216
On the same principle, the net commodity value outflow could be calculated as follows:
217
5!)* = !)* − !*)
218
where 687 expresses the net commodity value outflow from the NID to 27 other provinces,
219
87 expresses the commodity value outflow form the NID, and 8 7 expresses the commodity
220
value inflow to the NID.
+,
+,
,+
(7)
221
Based on the above EEBT model, we could analyze the characteristics of spatial-temporal
222
evolution for the net CEs-PT outflow between the NID and the other provinces from 1997 to 2007.
223
2.3.2 Modeling for driving force analysis—Two-stage SDA model. Based on the above
224
analysis, we applied structural decomposition analysis(SDA) approach to analyze the driving force
225
of the CEs-PT transference.
226
To get a general idea of SDA, we initially explored the CEs changes. Assume that there are
227
two time periods for which input-output data are available. Using superscripts 0 and 1 for the two
228
different years (0 earlier than 1, in our study, 1 express 2007 and 0 express 1997), our illustration
229
of structural decomposition in an input-output model focused on the change of net CEs-PT
ACS Paragon Plus Environment
Environmental Science & Technology
Page 10 of 33
230
outflow in two years. As usual, the net CEs-PT outflow in year t, n : (t = 0, 1), are found in an
231
input-output system as
232 233 234 235
5 = 5; × 5! " and 5< = 5;< × 5!C expresses the effect of related technology level change,
245
reflects the technology driving force, 1/26= +6=C ∆8> expresses the effect of the value
246
transference demand change, reflects the demand driving force. Formula 7 is the first stage of
247
SDA analysis model.
248
Based on the first stage of SDA analysis, we carried out the second stage of SDA analysis.
249
On the same principle, the change of net commodity value outflow in interprovincial trade can be
250
demonstrated by the following formula:
∆5!" = 5! " − 5!: is a column matrix with 30 rows.
263
2.3.3 Modeling for optimization—CR model. Focusing on the driving force of technology,
264
demand, structure and scale, we built a coupling relationship (CR) model for analysis the change
265
of ‘net commodity value outflow’ and ‘net CEs-PT outflow’ and the coupling relationship
266
between them. Based on the CR model, we could find out the optimization targets and formulate
267
the optimization policy.
268 269 270 271 272 273
In the CR model, We used the ‘contribution rate changes’ to measure ‘net commodity value outflow change’ and ‘net CEs-PT outflow change’. Thus we obtained the following:
∆5L! = 5L ! − 5L
(12)
Where ∆6QR and ∆6QS express the ‘contribution rate changes’ of ‘net commodity value
? outflow’ and ‘net CEs-PT outflow’, 8?T?UV = 8>? + 8? , means the sum of the commodity value ? outflow and inflow, ?T?UV = >? + ? , means the sum of the CEs-PT outflow and inflow. The 68?
274
and 6? express the net commodity value outflow and the net CEs-PT outflow of industry W in
275
year t, will be positive number for the net outflow industry, or the negative number for the net
276
inflow industry.
277
The whole application process of the CR model includes three steps:
278
Firstly, based on the Formula (12), we calculated the contribution rate changes of net
279
commodity value outflow (∆6QR ) and net CEs-PT outflow (∆6QS ) of each industry.
280
Secondly, based on the coupling relationship of ∆nPY and∆nPZ , we divided industries into 4
281
types and implemented different optimization policies for each industrial type according to the
282
industrial characteristics. Specific description is as follows:
ACS Paragon Plus Environment
Environmental Science & Technology
283 284 285
If ∆6QR >0 and ∆6QS 0 and ∆6QS >0, the
287
industry belongs to Type Ⅲ, characterized by net commodity value outflow has increased and net
288
CEs-PT outflow has also increased from 1997 to 2007. In Type Ⅲ, if ∆6QR >∆6QS >0, the industry
289
belongs to Type Ⅲ-A, characterized by net commodity value outflow has increased greater than
290
net CEs-PT outflow increased; if ∆6QS >∆6QR >0, the industry belongs to Type Ⅲ-B, characterized
291
by net commodity value outflow has increased less than net CEs-PT outflow increased. If ∆6QR 0;
460
and Type Ⅳ: ∆6QR
