APERC Workshop at EWG 46
Da Nang, Viet Nam, 18 November 2013
3. The APERC Macroeconomic Model
DU Bing, Researcher
Asia Pacific Energy Research Centre (APERC)
Contents
1
Background
2
Model Structure
3
Results and Discussions
2
1. Background
• The role of macroeconomic model in the APEC Energy Demand
and Supply Outlook
Industry demand model
Transport demand model
Residential & Commercial
demand model
Macroeconomic model
(Population, GDP, Savings rate,
Investment rate, Employment
rate, Education, etc. )
Electricity supply model
Other models …
3
Example of using macroeconomic model
Energy Intensity Data
Steel Production
Forecast
Steel Sector
Energy Demand
How can the macroeconomic model be
used to project steel energy demand?
GDP per capita,
Population …
Macroeconomic model
4
Why do we need a new macroeconomic model?
• Previously we used the IHS
Global Insight data as our
macroeconomic assumptions.
• Reasons not to use it anymore:
 We cannot explain (Models not available)
 Data not available for Brunei and PNG
 Some strange results (bias toward small economies such as
Singapore and Hong Kong)
 Expensive…
5
Why do we need a new macroeconomic model?
• There are currently many other macroeconomic projections.
However, it is difficult to use their results directly due to data,
document and source code availability, as well as time and
economy coverage problems.
Projections
Time coverage
Economy coverage
Detailed
Document
Database
Source
Code
Remarks
CEPII
1980-2050, annual
worldwide, 147 economies
O
O
O
using energy
as input
EIA
2006-2035, 5-year
intervals
22 selected economies
X
X
X
USDA
1969-2030, annual
worldwide, 190 economies
X
O
X
IMF
1980-2018, annual
worldwide, 188 economies
X
X
X
OECD
2011, 2030, 2060
42 selected economies
O
X
X
6
The model we are pursuing …
Based on proven, widely
adopted approach
Proven
Using open, published
research and
authoritative data
Open
sources
Simple
Easy to understand,
including only the
necessary elements
Accurate
Comprehensive
Covering all 21
APEC economies
Producing acceptable
results for long-term
projection
7
2. Model structure
Total GDP 𝑌 = 𝐴(𝑡)𝐾 𝛼 𝐿𝛽
Total Factor
Productivity,
TFP (A)
Catch-up
effect
Capital (K)
Education
effect
Savings rate
Income level
Income
growth rate
Economic
activity rate
Labor (L)
Investment rate
Culture,
institutions
and other
factors
Relationship
between
savings and
investment
rate
Population
structure
Economic
activity
rate of age
groups
Regional
differences
8
Three main factors in the model
• Capital accumulation is determined by investment rate (the share of investment
in GDP) and capital depreciation rate. investment rate is estimated based on the
relationship between savings rate and investment rate.
0.2500
• Labour is measured by the total economically
active population. For each age group, we have
population and economic activity rate data
from the ILO database.
0.2000
0.1500
0.1000
Population structure
Savings rate
0.0500
Investment rate
2012
2010
2008
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
0.0000
• TFP growth can be explained by a catchup effect, an education effect and an
interaction term between education and
catch up.
9
3. Results and Discussions
Projection of GDP per capita
90,000
GDP per capita (2005 USD PPP)
80,000
70,000
1980
2012
2035
60,000
50,000
40,000
30,000
20,000
10,000
0
10
Income growth trend
11
Total GDP growth rate
10.00%
9.00%
1980-2012
2012-2035
Annual GDP growth rate
8.00%
7.00%
6.00%
5.00%
4.00%
3.00%
2.00%
1.00%
0.00%
12
Comparison with the CEPII projection
120,000
GDP per capita (2005 USD PPP)
100,000
APERC 2035
CEPII 2035
80,000
60,000
40,000
20,000
0
13
Thanks for your attention!
du.bing@aperc.ieej.or.jp
14