B
A
S
I
C
Task 1:
Energy Models in China
Fei TENG
Global Climate Change Institute,
Tsinghua University
B A S I C
Outline




The Time Line of Energy Models
Overview of Bottom-up Models
Overview of Top-down Models
Major Findings from The Task
B
A
S
I
C
Time Line of Energy Models
Bottom-up
1980
MARKAL IPAC LEAP
1990
Top-down: Input-Output
3E
2000
YE’s
2005
Liang’s
PRCGEM HE
TEDCGE
Computable General Equilibrium
DRCSC’ CNAGE IPAC-SGM
B
A
S
I
C
Overview of Bottom-up Models
Reference Scenario
Name
Object
optimization of
MARKAL energy system path,
emissions forecasting
optimization of
LEAP
energy system path,
emissions forecast
Based on the Asian
center to describe the
AIM
problem and policy
analysis
optimization of
3E
energy system path,
emissions forecasting
Base
Year
GDP
Energy
Emission
1995
1995-2030 1995-2030
6.45%
2.4%
1655MtC
(2030 BAU)
1999
1999-2030 1999-2030
6.6%
2.6%
1700 MtC
(2030 BAU)
---
---
2000
2000-2050
2000-2030
6.2% (from
2.8%
MEM)
---
---
B
A
S
1710 MtC
(2030 BAU)
I
C
MARKAL Model







Developed by BNL and KFA, a multi-periods LP model
Minimize the cost of satisfying the energy demand
China MARKAL-MACRO Model, Base Year: 1995.
Period covered: 1995-2050, every 5 years.
Sectors: 5 Sectors, Agriculture, Industry, Residence, Transport and
Service, 30 sub-sectors in detail.
Technology: 20 energy carriers; 36 reference technologies and 35
advanced technologies.
Applications:
•
•
INET: impacts of emission reductions on China’s GDP with MARKAL-MACRO
Model; Beijing’s energy supply scenarios and possible impacts;
SHESRI: responses of the energy system to energy structure adjustment
policies in Shanghai
B
A
S
I
C
MARKAL Model

BAU: 2000-2050
•
•
Population: 1.294 billion for 2000; 1.495 billion for 2020;
1.56 billion for 2030; 15.75 billion for 2050.
Economy Growth: 1074.6 billion $ for 2000; 3710.5
billion $ for 2020; 6338.1 billion $ for 2030 (1995 Price)
•
Economy Structure.
•
Energy Service: 5 sectors.
•
•
Resource constrain for primary energy: e.g. Coal,
1823Mt for 2010, 2512Mt for 2030.
Energy Technologies: Reference technologies +
Advanced technologies (CCS + polygeneration)
B
A
S
I
C
MARKAL Model

Scenarios:




ADV: considering advanced technologies.
C20P, C30P and C40P: reduce emission
from 2020. 2030 and 2040.
N1 and N2: constrain on Nuclear capacity.
Four reduction scenarios: 11%, 23%,
27.4% and 46.4%.
LEAP Model








Base Year: 1999.
Period covered: 1999-2030.
Sectors: 5 Sectors, Agriculture, Industry, Residence, Transport and
Service.
Scenario-based model describing the production, transformation and
consumption of energy
No linkage between energy price and the economy
Activity-based energy demand forecasting
Used for environmental impact analysis in conjunction with
Technology and Environmental Database
Application
•
NDRC Energy Research Institute (ERI), INET and SHESRI
•
ERI: Sustainable Energy Development Scenarios in China
•
INET: China’s energy system under future Northeast Asia
cooperation scenario
LEAP Model
Scenarios
Results
ERI:
In 2020, S1, S2 and S3 will reach 3100Mtce,
S1: focus on the energy efficiency
2761Mtce, 2318Mtce separately.
increment by economic
The total carbon emissions for three scenarios will
development
reach 1899.9Mtc，1659 Mtc，1265.3 Mtc
S2: focus on the optimistic
separately.
sustainable development and
energy development scenario
S3: is the ideal scenario.
INET :
BAU scenario
H-E scenario (natural gas
import)
H-I scenario (nuclear,
renewable energy)
Calculation results show that the primary energy
demand in three scenarios (BAU, H-E, H-I
respectively) can reach 2967、2842、3119Mtce
respectively, with import taking up 21%、31%、
13% respectively of the total primary energy
supply in these scenarios. CO2 emission will
reach 6236、5568、5568Mt respectively and
SO2, NOx emission will amount to 30.2、21.6、
27Mt and 23、17、21Mt respectively.
B
A
S
I
C
AIM Model





Detailed technical assessment model for evaluation of
technical and GHG mitigation policies
Suitable for short- and medium-term analysis
Could be used to evaluate the effects of one single or
several policies
Application: ERI
Insufficiencies:
•
•
•
No linkage with economic model and exogenous setting of
energy demand
Social and other barriers not considered
Not able to cover all technologies
Economy-EnergyEnvironment（3E）Model




Three components: macroeconomic model
(MEM), end use forecasting model (EDFM),
energy system optimization model (ESOM)
MEM: to estimate the long-term economic
development
EDFM: to forecast end use energy demand with
energy intensity index, elastic coefficient method
and econometrics method
ESOM: to optimize the energy system based on
energy flow networks
B
A
S
I
C
Economy-EnergyEnvironment（3E）Model


Developed by INET, Tsinghua
Application:
•
•
•

China’s CO2 abatement cost during 2015-2030
Relationship between China’s CO2 abatement cost and
energy strategies
Responses of energy system to the mitigation policies
Insufficiencies
•
•
•
Outputs of ESOM are technology-based, not sectorbased
MEM model is a macro-econometric, not suitable for
long-run forecasting
Open-Loop, no feed back
Major Findings from Bottomup Models


Difficult to compare the result from
different models because of different
scenarios setting and period covered.
A set of scenario should be developed


Social economic scenario: population, GDP
etc.
Energy Service scenario: agriculture,
industry, household, commerce and
transport.
Overview of Top-down Models
Name
YE
Liang
HE
Structure
Model Elements
Economy
Dynamic inputIO table
output model
Input-output analysis
IO table
+ scenario analysis
Emission
Technology Climate
Yes
IO coefficient
and sub-sectors
Yes
No
IO coefficient
No
Static CGE
CGE
Fossil fuel combustion CES
Yes
PRCGEM Static CGE
CGE
Fossil fuel combustion Leontief/CES
Yes
DRCSC’s
Dynamic CGE
CGE
No
TEDCGE
Dynamic CGE
CGE
No
CES
Fossil fuel combustion
CES
and cement production
CNAGE
Static CGE
CGE
Fossil fuel combustion CD
Yes
IPAC-SGM SGM CGE
CGE
---
Yes
---
B
A
S
Yes
I
C
YE’s IO Model





Developed by INET
Multi-sectors, multi-period nonlinear input-output
model
Maximize the aggregated consumer utility while only
the utility caused by consumption is considered
There GHG emission sources considered: combustion
of fossil fuel, production process and some byproducts
Application: impacts of different mitigation scenarios
on the GDP
B
A
S
I
C
Liang’s IO Model

Application
•
•
•
China’s energy demand and GDP energy
intensity during 2010-2020
Impacts of different socio-economic factors on
energy demand and energy intensity
Insufficiencies
•
RAS adjustment method no suitable for a fast
developing economy like China
B
A
S
I
C
Comparison between IO
models
Developer
Descriiption
Objective
YE
Data source: 1987 33*33 sectors input-output
table.
Maximize the utility function of a
representative consumer. The consumers
in INET are divided into urban residents
and rural residents.
Some sectors in input-output table is divided
into more detailed sub-sectors to show
the substitution and complementary
among different factors according to the
status of different technologies.
To compare different abatement
schemes and calculate the
optimal reduction path
under given reduction goal.
Liang
Data source: 1997 40*40 sectors input-output
table, other data came from ‘National
communiqué 1997 on national economic
and social development in China’
Scenarios setting: 6 scenarios are considered
Year 1997 is set as the base
year in this model. Energy
demand and energy
intensity are predicted using
this model.
B
A
S
I
C
HE Model





A static model with 9 sectors
Impacts of carbon tax on the economy
Carbon tax imposed on the production and import of
fossil energy, and thus no consideration of CO2
emission from other sources
China’s CDM potential: adjustment of economic
structure, technological progress, adjustment of
energy structure and energy efficiency improvement
1997 I-O table
B
A
S
I
C
PRCGEM




Large-scale model with 118 sectors and 30 areas
Mainly impacts of trade liberalization policies, but also
of environmental policies
Carbon tax imposed on the consumption of fossil fuel,
and thus no consideration of CO2 emission from other
sources
Long-term and short-term mitigation cost, with
different assumptions
B
A
S
I
C
DRCSC’s CGE Model


Multi-sector dynamic model
Application
•
•
•
•
•
the future trend of industry structure in China
the environment impact of these trends
impact of pollution limitation policy on the economic
growth and industrial structure
environment impact of trade liberalization and
globalization in China
No consideration of CO2 mitigation
B
A
S
I
C
TEDCGE Model




10 sectors
Carbon tax on fossil fuel production and import, based
on adjusted factors
Emission from industrial process also considered
Impacts of carbon tax under four scenarios:
•
•
•
•

Only carbon tax
Carbon tax and whole transfer payment
Carbon tax and 50% transfer payment
Carbon tax and tax reduction in other areas
China’s CO2 mitigation potential and cost
CNAGE Model


Developed jointly by Chinese and Norwegian
Statistic Bureaus
Impacts of carbon tax of two levels on the shortand long-term GDP and productions of different
sectors
B
A
S
I
C
IPAC-SGM Model



Developed jointly by ERI and US Pacific Northwest
Laboratory
20 sectors with 9 energy production and 11 energy
consumptions sectors
Application
•
•
China’s energy scenarios till 2050
Impacts of carbon tax, technology investment and
technology cooperation on GHG emissions, the economy and
different sectors
B
A
S
I
C
Comparing CGE models
Name
Descriiption
Objective
HE
Static CGE model
9 sectors, CES production function with (K,L)E fashion.
1997 input-output table
Unit carbon tax collected on the fossil fuel production.
Armingtion assumption and small country assumption.
To analyze the impacts of
carbon tax on the
national economy
PRCGEM
Dynamic CGE model.
118 sectors version and 34 sectors version.
CES production function with (K,L)E fashion, one
representative consumer.
Data basis: the 1992 input-output table. Only the emission
caused by fossil fuel combustion is considered.
A unit tax on the carbon content in fossil fuel instead of
carbon dioxide emission.
To analyze the impacts of
trade liberalization.
Carbon tax also can be
analyzed when energy
is included as a
productive factor.
DRCSC’s
B
A
To analyze the impacts of
globalization and
trade liberalization on
environmental
pollutions.
S
I
C
Comparing CGE Models
TEDCGE
Dynamic CGE model.
10 productive sectors, CES production function with (K,E)L fashion.
Two reprehensive consumers: urban resident and rural resident.
The Armington assumption and small country assumption.
An eclectic treatment for emission based on the carbon content of
fuel, the fraction of stored carbon, the fraction of carbon oxidized.
The carbon tax is collected on the production and import of the
primary energy. Also includes the carbon dioxide emission from the
industrial process, especially the cement production.
To analyze the
impacts of carbon
tax on the national
economy
CNAGE
Static CGE model.
Constant return to scale Cobb-Douglas production function with the
aggregation fashion of (L, E, K).
19 commercial energy products while 5 types for final consumption.
The labor market is not clear for the unemployment and determined
by the profit maximization behavior of producers.
To analyze the
impacts of carbon
tax on the national
economy
IPACSGM
The economy system includes residential sector, government sector,
agriculture, energy and other sectors.
Two kinds of labor: urban labor and rural labor to show the cost
difference between these two labors.
The investment for nuclear power and hydro power is under the
control of central government.
To analyze the
impacts of carbon
tax on the national
economy and
possible strategy
for abatement.
Major Findings from CGE models




Elasticity estimation to reduce uncertainties
Model structure: competitive market
Detailed expression of technology in the models
Treatment of non-commercial energies
B
A
S
I
C
Suggestion





Strengthened fundamental econometric research to
complete the data basis and reduce the impacts of
estimated parameters
More comprehensive welfare analysis for different
consumer groups
More attention given to specific market conditions and
price regulation, and thus some structural CGE models
could be the future direction
Non-commercial energies be considered
More detailed description of technologies as
technology plays an essential role in making relevant
decisions
C
B A S
I
Future Works


Compare not only the model structure
but also the scenarios.
Reorganize the material to account for
different audiences.
Thank for your attention
B
A
S
I
C