Transcript Tol - Indico

Earth System Economics
Richard S.J. Tol
Hamburg, Vrije & Carnegie
Mellon Universities
The Vision
• A predictive Earth System Model, that is,
a model that dynamically links all major
components of the earth system, and
forecasts their behaviour with only current
and past observations, that is, without
scenarios
• This ESM should have current decisions as
input, so that it can be used for policy
advice
Whose Vision?
• After successfully coupling atmosphere,
ocean and ice models, the CGCMs are
thirsting for more, turning their immediate
attention to terrestrial vegetation, lower
trophic levels in the ocean, and
atmospheric chemistry
• The „human dimension“ is lower down the
list of priorities but will be steadily moving
upwards, hitting the top in 5-10 years
Why Do We Care?
• The ESMs will bring pots of money and
influence
• Initially, they will want interactive
greenhouse gas emissions and land use
scenarios, followed by detailed spatial
patterns of other emissions to air, and
variability
• For some of these things, we have some
theoretical insight, but no applied models,
and for others we have little clue: There is
an intellectual challenge as well
Intermediate Steps
• We already have many simple ESMs, called
integrated assessment models
• Many IAMs are „conditionally predictive“
partly coupled systems, but the crude
spatial and temporal resolution makes many
problems disappear
• People are now moving towards IAMs „of
the next generation“ or ESMs of
intermediate complexity
Population
Fertility
Mortality
Migration
Economic System
Economy
Energy
Other gases
Interfaces
Water use
Land use
Recreation
Impacts
Sea level rise
Health
Tourism
Energy demand & supply
Phys-Chem System
Ocean carbon cycle
Climate
Other gas cycle
Biogeochem System
Hydrology
Vegetation
Agriculture
Forestry
Immediate Steps
• One field, in which I see short-term
benefits as well as opportunities for longterm learning, is higher-order impacts of
climate change
• To date, most economic impact studies are
based on direct costs estimates, and a few
on partial equilibrium models
• We clearly can do better than that and
study the general equilibrium effects, the
structural effects, and the growth effects
Growth
• Consider a Ramsey-Cass-Koopmans growth
model (maximise utility, accumulate capital,
labour and technology exogenous)
• Climate affects utility (no effect), output,
depreciation, and labour
• Keep savings constant for the moment
• Less output implies less capital
accumulation
• Faster depreciation implies less capital
• Less labour implies more capital per worker
Growth -2
• Now make savings endogenous
• Less output implies a lower return on
capital
• Less labour implies less need to invest
• Faster depreciation may implies more
savings to make up or less to reflect the
reduced returns
• However, net savings are unambiguously
down
• Capital effect is negative unless black
plague
0.25
0.2
0.15
No Climate Change
Solow
Ramsey
Romer
Mankiw
1.4
1.2
1
percent
fraction of GDP
1.6
Mankiw
Romer
Ramsey
Solow
Direct Cost
Ramsey, depreciation
0.1
0.8
0.6
0.4
0.05
0.2
0
0
2005
2025
2045
2065
2085
2105
2125
2145
2165
2185
2205
2005
2025
2045
2065
2085
year
2125
2145
2165
2185
2205
year
1.6
1.6
No climate change
1%
5%
10%
15%
1.4
1.2
1.2
1
1
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0
No Climate Change
1% Damage
5% Damage
10% Damage
15% Damage
1.4
percent
percent
2105
0
2005
2025
2045
2065
2085
2105
-0.2
2125
2145
2165
2185
2205
2005
2025
2045
2065
2085
2105
-0.2
year
year
2125
2145
2165
2185
2205
General Equilibrium
• Most studies estimate the direct costs of
climate change, that is, price times
quantity
• This is a crude welfare measure
• CGEs also estimate welfare change, so that
we cannot use direct cost studies as inputs
to CGE modelling
• Inputs to CGEs can be changes in
endowments, productivity, and demand but
the last is a bit more complicated
Implementing Climate-Change Impacts on Health in GTAP-EX
Health impacts:
selected results 2050
2050
USA
EU
EEFSU
JPN
RoA1
EEx
CHIND
RoW
Equivalent
Variation
Gdp
Co2 Emissions
($ US million)
(% change w.r.t.
baseline)
(% change w.r.t.
baseline)
-5052.14
-4227.32
-795.954
-649.416
-751.162
-464.824
-553.109
-822.113
-0.02
-0.018
-0.023
-0.006
-0.015
-0.003
-0.01
-0.007
-0.005
-0.004
-0.007
0.001
-0.004
0
0.001
0
Implementing Climate-Change Impacts on Sea Level Rise in GTAP-EX
Sea-level rise:
selected results 2050
2050
USA
EU
EEFSU
JPN
RoA1
EEx
CHIND
RoW
Equivalent
Variation
Gdp
Co2 Emissions
($ US million)
(% change w.r.t.
baseline)
(% change w.r.t.
baseline)
-785.325317
-841.421326
-93.030487
88.591309
30.607679
-566.428589
-410.882568
-726.667908
-0.001263
-0.001301
-0.000972
-0.000424
-0.000270
-0.009059
-0.013259
-0.008243
0.002692
0.000811
0.004347
0.014146
0.004597
-0.003644
-0.002694
-0.009216
The Immediate Challenge
• Land and water use are heavily affected by
the climate, and affect the climate in
return; the spatial pattern matters
• Land and water markets are distorted,
regulated and location-specific package
deals
• I think a lot can be done with spatial
equilibrium models and risk-averse
farmers; power weights in bargaining
games; and Krugman‘s new economic
geography, and ideas are being tested
Longer-Term Challenges
• The natural science components of Earth
System Models have both short-term and
long-term dynamics; variability and
equilibrium
• The same thing will be asked from social
science components
• Complex models are tested by reproducing
the past – economic historians are still, by
and large, data collectors, but their
progress is amazing
Conclusions
• There will be a market for ESM-compatible
economic models
• At the moment, no economic model supplies
to that niche, but with a number of
adjustment this can be arranged
• Filling the niche requires quite a bit more
in theoretical and applied economics; and
prospects are uncertain because of the
differences in space, time and closure