Transcript to a pps diaporama

WITCH Model Description and Applications
FEEM
The WITCH Team:
Andrea Bastianin
Valentina Bosetti
Carlo Carraro
Enrica De Cian
Alice Favero
Emanuele Massetti
Lea Nicita
Elena Ricci
Fabio Sferra
Massimo Tavoni
1
WITCH Model, Description and Applications
www.feem-web.it/witch
The WITCH Model: An Introduction
2
WITCH Model, Description and Applications
The WITCH Model
WITCH: World Induced Technical Change Hybrid model
Hybrid I.A.M.:
 Economy: Ramsey-type optimal growth (inter-temporal)
 Energy: Energy sector detail (technology portfolio)
 Climate: Damage feedback (global variable)
 12 Regions (“where” issues)
 Intertemporal (“when” issues)
Economic Activity
 Game-theoretical
set-up (free-riding
incentives)
temperature



3
Atmosphere
Biosphere
Bosetti V., E. De Cian, A. Sgobbi and M. Tavoni (2009). “The 2008
WITCH Model: New Model
emissions
Features and Baseline,” FEEM Working
Paper October 2009.Deep Oceans
Bosetti V., E. Massetti, M. Tavoni (2007). “The WITCH Model, Structure, Baseline, Solutions”, FEEM
Working Paper 10.2007.
Bosetti,
V., C. Carraro,
Energy
Use M. Galeotti, E. Massetti and M. Tavoni (2006). “WITCH: A World Induced
Technical Change Hybrid Model”, The Energy Journal, Special Issue. Hybrid Modeling of EnergyEnvironment Policies: Reconciling Bottom-up and Top-down, 13-38.
WITCH Model, Description and Applications
The WITCH model - http://www.feem-web.it/witch/
A hybrid energy-economy-climate model
 Scale: global, with the world divided in 12 regions
 Economy: top-down intertemporal optimal growth model,
dynamic, perfect foresight
 Energy: bottom-up description of technological options:
 Electric and Non Electric energy use
 Six fuel types specified (oil, gas, Coal, Uranium, traditional
and advanced biofuels)
 Seven technologies for electricity generation
 Endogenous technical change – Learning-By-Doing and
Learning-By-Researching
 Climate: damage feedback via temperature change
 Strategic: non cooperative interactions between region with
externalities (environmental, price of exhaustible resources,
technological spillovers, and trade of emission permits)
4
WITCH Model, Description and Applications
Distinguishing Features
Bottom-up characterisation of the energy sector
Detailed representation of technological change
 Learning-By-Doing in W&S
 Energy intensity R&D
 Breakthrough Technologies (two factors learning curves)
Several channels of interactions among regions
 Technological spillovers
 Environmental externality
 Exhaustible common resources (coal, natural gas and uranium)
 Trade of emission permits
 Trade of oil
Game-theoretic set-up makes it possible to model strategic behaviour (open loop
Nash game) and to describe cooperative and non-cooperative solutions
5
WITCH Model, Description and Applications
Two possible regional aggregations
World countries, aggregated into 12 regions












6
United States (USA)
Western EU countries (WEURO)
Eastern EU countries (EEURO)
Canada, Japan and New Zealand
(CAJANZ)
Korea, Australia and South Africa
(KOSAU)
Non-EU Eastern European
countries, including Russia (TE)
Latin America, Mexico and
Caribbean (LAM)
Middle East and North Africa
(MENA)
South Asia, including India (SASIA)
China, including Taiwan (CHINA)
Sub-Saharan Africa excluding
South Africa (SSA)
South East Asia (EASIA)
WITCH Model, Description and Applications












United States (USA)
Western EU countries (WEURO)
Eastern EU countries (EEURO)
Canada, Australia and New
Zealand (AUCANZ)
Korea, Japan (JPNKOR)
Non-EU Eastern European
countries, including Russia (TE)
Latin America, Mexico and
Caribbean (LAM)
Middle East and North Africa
(MENA)
South Asia, including India (SASIA)
China, including Taiwan (CHINA)
Sub-Saharan Africa including South
Africa (SSA)
South East Asia (EASIA)
The Objective Function and Budget Constraint
For each region (n) forward-looking central planner maximizes present value of (log) per
capita consumption (5-yr time steps):
(1)
W (n)   L(n,t)  log  c(n,t)   R(t)
t
choosing the optimal path of investment variables simultaneously and strategically with
respect to the other decision makers

Consumption of the single final good obeys to the economy budget constraint:
GDP
(2)
7
Final
Good
Energy
R&Ds
Electricity
Generation
Operation &
Maintanance












C
n
,
t

Y
n
,
t

I
n
,
t

I
n
,
t

I
n
,
t

O&
n
,
t



C
R
&
D
,
j
j
j
j
j
j









P
n
,
t
X
n
,
t

P
n
,
t
CCS
n
,
t

f
CCS
ff
Net fuel
expenditures
WITCH Model, Description and Applications
CCS (Transport and
storage costs)
Output and Climate Damage
Gross output is produced combining the inputs capital, labour (=population) and
energy services using a nested, Constant Elasticity Production Function
 






GROSS GDP



(3) Y












n
,
t

TFP
n
,
t
(
n
)

K
n
,
t
L
n
,
t

(
1

(
n
))

E
n
,
t

n
,
t
1

(
n
)(
n
)
C
1
/
Climate change damage is a non-linear function. Climate change impacts can be
either positive or negative and they are region-specific
(4)
8


2

(
n
,
t
)

1

T
(
t
)

T
(
t
)
1
,
n
2
,
n
WITCH Model, Description and Applications
Output and Climate Damage
Net output is obtained after subtracting expenditure for fossil fuels, which is considered as
a net loss for the economy
CCS is the amount of CO2 captured from the atmosphere and PCCS the corresponding
costs that the economy has to pay to external suppliers of CCS know-how
(5)
9

n, t   (1   (n))  ES n, t 
n, t 
  f  Pf n, t  X f ,extr n, t   Pfint t  X f ,netimp n, t  
 PCCS n, t  CCS n, t 
Ynet n, t  

TFPn, t   (n)  K C
WITCH Model, Description and Applications
1  ( n )
n, t  L
 (n)



1/ 
Production Tree and Energy Technologies
Production nest and the elasticity of substitution
Legenda: KL= Capital-labour aggregate; K = Capital invested in the production of final good; L =
Labour; ES = Energy services; HE = Energy R&D capital; EN = Energy; EL = Electric energy;
NEL = Non-electric energy; OGB = Oil, Backstop, Gas and Biofuel nest; ELFF = Fossil fuel
electricity nest; W&S= Wind and Solar; ELj = Electricity generated with technology j (IGCC plus
CCS, Oil, Coal, Gas, Backstop, Nuclear, Wind plus Solar); TradBiom= Traditional Biomass;
TradBio= Traditional Biofuels; AdvBio= Advanced Biofuels
10
WITCH Model, Description and Applications
Electricity Production - 1
Electricity is obtained by combining in fixed proportions the installed power generation
capacity (K), operation and maintenance equipment (O&M) and fuel resources
consumption (X) (when needed)
Power Plant
Fuels
Operation and
Maintenance
11
WITCH Model, Description and Applications
Electricity
Production function are characterized by region-specific
parameters that account for the technical features of each
power production technology, such as the low utilisation
factor of renewables, the higher costs of running and
maintaining IGCC-CCS and nuclear plants
Electricity Production - 2


Electricity production is described by a Leontief production function


(6) EL







n
,
t

min
K
n
,
t
;
O&
n
,
t
;
X
n
,
t
j
n
,
j
j
n
,
j
j
j
j
,
EL
μ translates power capacity into electricity generation
Τ differentiates O&M over technologies
ζ yields the quantity of fuels needed to generate 1 KwH of electricity
Power Generation capacity (Power Units) depends on cumulated investments (I) and
investments costs (SC) which are time and region-specific:
(7)

I
(
n
,
t
)
j




K
n
,
t

1

K
(
n
,
t
)
1


j
j
j
SC
(
n
,
t
)
j
WITCH Model, Description and Applications
12
Technical Change – Learning-By-Doing
Endogenous Technical Change (ETC) accounts for the accumulation of both:
• Experience (Learning-By-Doing)
• R&D investment (Learning-By-Researching)
Learning-By-Doing via experience curves in power plants investment cost
(9)

2
j




SC
n
,
t

B
K
n
,
t

1



j
j
j
n
tn

l
P
og
R
World learning, assuming full technology spillover: investments in additional capacity
by virtuous regions drive down investment costs worldwide, with benefits also for the
non investing regions
13
WITCH Model, Description and Applications
Technical Change – Energy Efficiency
Learning-By-Researching via energy R&D increasing energy efficiency (Popp, 2004)
(10)
ESn, t   
H HE(n, t )

  EN EN(n, t )

 1/ 
The R&D sector exhibits intertemporal spillovers and the production of new "ideas"
follows an innovation possibility frontier (Popp, 2002; Jones,1995):
(11)
Z n, t   anI R&D (n, t ) b HE(n, t ) c
The flow of new ideas adds to the previously cumulated stock and generates the
total amount of knowledge available to country n at time t:
(12)
14
HE(n , t  1 )  HE ( n, t )(1   R & D )  Z n, t 
WITCH Model, Description and Applications
Technical Change – International Spillovers
The R&D sector exhibits also international knowledge spillovers:
(13)
Z n, t   an I R&D (n, t ) b HE(n, t ) c SPILLn, t 
d
The contribution of foreign knowledge to the production of new domestic ideas
depends on the interaction between two terms: the first describes the absorptive
capacity whereas the second captures the distance from the technology frontier,
which is represented by the stock of knowledge in rich countries
(USA, WEURO, EEURO, CAJANZ and KOSAU)
(14)
HE
(
n
,
t
)
SPILL
(
n
,
t
)

(
HE
(
n
,
t
)

HE
(
n
,
t
))

HE
(
n
,
t
)
HI

HI
Absorptive capacity
WITCH Model, Description and Applications
Distance from the frontier
Technical Change – Advanced Biofuels
Learning-By-Researching via dedicated R&D decreasing the cost of the cellulosic

biofuels, PADVBIO(n,t)
(15)





P
n
,
t

P
n
,
0

(
TOT
(
n
,
t
))
ADVBIO
ADVBIO
R
&
D
,
ADVB
where  stands for the relationship between new knowledge and cost
t
(16)

TOT
(
n
,
t
)

K
(
n
,
t

2
)

I
(
n
,
)


R
&
D
,
ADV
R
&
D
,
ADV
BIO
R
&
D
,
ADV
BIO
n


t

1
The stock of world R&D (ΣK) accumulates with the perpetual rule and it will influence
other regions with a 10-year (2model periods) delay. The time lag is meant to account
for the advantage of first movers in innovation
16
WITCH Model, Description and Applications
Technical Change – Breakthrough Technologies
Learning-By-Doing and Learning-By-Researching via cumulative capacity and
dedicated R&D decreasing the cost of breakthrough technologies, following a two
factors learning curve

c
(17)

b


P
R
&
D
CC
t,
ec
T 
t,
ec
T

2
t,
ec
T

*







P
R
&
D
CC
t,
ec
0 
t,
ec
0
t
,
ec
0


where the R&D stock (R&D tec) accumulates with the perpetual rule and it is also
augmented by the stock of R&D accumulated in other regions through a spillover effect,
SPILL, similarly to energy efficiency R&D
Two breakthrough technologies: one as substitute for nuclear in power generation and one
as substitute for oil in the non-electric sector (transport)
17
WITCH Model, Description and Applications
Major Research Topics
Mitigation options and costs
Innovation
Uncertainty
International policy architectures and coalition theory
Optimal balance between mitigation and adaptation
18
WITCH Model, Description and Applications
Major Research Topics
Mitigation options and costs
Innovation
Uncertainty
International policy architectures and coalition theory
Optimal balance between mitigation and adaptation
19
WITCH Model, Description and Applications
Mitigation Options, Technologies, Carbon Markets - 1
Major Areas of Research:









20
Optimal investments in energy technologies
Optimal investments in R&D
Climate policy costs: global and distribution
Climate policy costs with limits on the penetration of carbon free
technologies
Modeling backstop technologies
Investments in electricity grids
International trade of oil
Financing climate policy
Carbon markets
WITCH Model, Description and Applications
Mitigation Options, Technologies, Carbon Markets - 2
Key Findings:
 First energy efficiency, then decarbonization
 Climate policy costs are moderate for a 650 ppm CO2-eq
 Climate policy costs increase but are still reasonable for a 550 ppm
CO2-eq scenario
 No silver bullet. Complex portfolio mix with: nuclear, renewables,
coal with ccs
 Stringent climate policy is unfeasible with delayed (2030) or
incomplete action (China, India)
 Modeling international trade of oil tilts distribution of costs towards
oil exporting countries
21
WITCH Model, Description and Applications
Changes in Energy and Carbon Intensities
Energy savings and efficiency should be pursued vigorously in the short term, but
decarbonisation is essential from 2030 onwards already
100%
Decarbonization
80%
2100
450
550
60%
2050
40%
2030
20%
0%
0%
2030
2030
2050
20%
40%
550
650
BAU
2100
past 30 yrs
2050
2100
60%
80%
-20%
Energy Intensity Improvement
22
WITCH Model, Description and Applications
100%
Electricity Mix, 550 ppm CO2-eq
World Electricity Generation Shares
1
0.9
0.8
0.7
Nuclear
Hydroelectric
Oil
Gas
IGCC+CCS
Trad Coal
Wind&Solar
0.6
0.5
0.4
0.3
0.2
0.1
0
2000
23
2020
WITCH Model, Description and Applications
2040
2060
2080
2100
Mitigation Options, Technologies, Carbon Markets - 3
References:
 Bosetti, V., C. Carraro, E. Massetti, A. Sgobbi and M. Tavoni (2009). “Optimal
Energy Investment and R&D Strategies to Stabilise Greenhouse Gas
Atmospheric Concentrations,” Resource and Energy Economics, 31(2): 123137.
 Bosetti, V., C. Carraro and E. Massetti (2009). “Banking Permits: Economic
Efficiency and Distributional Effects,” Journal of Policy Modeling, 31(3): 382403.
 De Cian, E. and M. Tavoni (2009). “Sharing the burden to 2050: what role for an
international carbon market?” Fondazione Eni Enrico Mattei, July 2009, mimeo.
 Bastianin, A., A. Favero and E. Massetti (2009). “Investing in a Low-Carbon
World,” Fondazione Eni Enrico Mattei, July 2009, mimeo.
 Massetti, E. and F. Sferra (2009). “A Numerical Analysis of Optimal Extraction
and Trade of Oil Under Climate Policy and R&D Policy,” Fondazione Eni Enrico
Mattei, July 2009, mimeo.
 Tavoni, M., B. Sohngen and V. Bosetti (2008). "Forestry and the Carbon Market
Response to Stabilize Climate", Energy Policy, 35: 5346-5353.
24
WITCH Model, Description and Applications
Major Research Topics
Mitigation options and costs
Innovation
Uncertainty
International policy architectures and coalition theory
Optimal balance between mitigation and adaptation
25
WITCH Model, Description and Applications
Innovation - 1
Major Areas of Research:
 Directed technical change
 Human capital accumulation
 International knowledge spillovers
 Intersectoral knowledge spillovers
 Two factors learning curves for backstop technologies
26
WITCH Model, Description and Applications
Innovation - 2
Key Findings:
 Sharp increment of energy R&D (four-fold) is needed
 R&D investments in backstop technologies play a key role when
there are constraints to the development of nuclear and/or
renewables
 Modeling international disembodied R&D spillovers does not
change mitigation policy costs
 Intersectoral R&D spillovers might have a greater influence
 With directed technical change, overall R&D investments
decline with climate policy, and GDP losses increase
 Human capital is pollution-using (due to the complementarity
between labor and energy) and therefore climate policy redirects investments away from education toward R&D which
instead is pollution-saving
27
WITCH Model, Description and Applications
% of world GDP
Investment in R&D with Breakthrough Technologies
0.14
Baseline
550ppm
0.12
550ppm w ith backstops
0.10
0.08
0.06
0.04
0.02
0.00
2007
2012
2017
2022
2027
2032
2037
2042
2047
2052
 Breakthrough technologies can only become available with
substantial investments in R&D
 Energy R&D expenditures increase up to 0.12% of GDP, vs.
0.02% in the BAU
28
WITCH Model, Description and Applications
500
450
550ppm
400
 The price of carbon is much lower with
breakthrough technologies
550ppm w ith
backstops
350
300
250
 Crucial role to decarbonize non-electric
energy (transport)
200
150
100
50
0
2007
2012
2017
2022
2027
2032
2037
2042
 And therefore the costs of
stabilisation are much lower,
especially in the long term
2047
2052
% change in GDP with respect to baseline
$US/tCO2 eq
Mitigation Costs with the Backstop Technologies
0.0
-1.0
-2.0
-3.0
-4.0
-5.0
-6.0
550ppm w ith backstops
-7.0
29
550ppm
WITCH Model, Description and Applications
-8.0
2007
2012
2017
2022
2027
2032
2037
2042
2047
2052
2057
2062
2067
2072
2077
2082
Induced Technical Change and GWP Losses
With respect to a Full Induced Technical change (ITC) Scenario Gross World
Product (GWP) losses are:
Overestimated when there is no
ITC in the Energy Sector
No ITC Energy
Understimated when there is no
ITC in the Non-Energy Sector.
Underestimated when there is
no ITC
Understimated when there is
only exogenous crowding out
of Non-Energy R&D
30
WITCH Model, Description and Applications
Full ITC
No ITC
Exogenous
Crowding-out
No ITC Non-Energy
3.4%
3.5%
3.6%
3.7%
3.8%
3.9%
4.0%
Discounted GWP Loss from Climate Policy (%)
4.1%
Innovation - 3
References:
 Carraro, C., E. Massetti and L. Nicita (2009). “How Does Climate Policy Affect
Technical Change? An Analysis of the Direction and Pace of Technical Progress
in a Climate-Economy Model.” The Energy Journal, Forthcoming.
 Bosetti, V., C. Carraro and M.Tavoni (2009). “Climate Policy after 2012.
Technology, Timing, Participation,” CESifo Economic Studies, Forthcoming.
 Bosetti, V., C. Carraro, E. Massetti, A. Sgobbi and M. Tavoni (2009). “Optimal
Energy Investment and R&D Strategies to Stabilise Greenhouse Gas
Atmospheric Concentrations,” Resource and Energy Economics, 31(2): 123137.
 Bosetti, V., C. Carraro, R. Duval, A. Sgobbi and M. Tavoni (2009). “The Role of
R&D and Technology Diffusion in Climate Change Mitigation: New Perspectives
using the WITCH Model.” OECD Working Paper No. 664, February.
 Carraro, C., E. Massetti and L. Nicita (2009). “Optimal R&D Investments and the
Cost of GHG Stabilization when Knowledge Spills across Sectors.” Fondazione
Eni Enrico Mattei, July 2009, mimeo.
 Carraro, C., E. De Cian and M. Tavoni (2009). “Human Capital Formation and
Global Warming Mitigation: Evidence from an Integrated Assessment Model.”
Fondazione Eni Enrico Mattei, July 2009, mimeo.
31
WITCH Model, Description and Applications
Major Research Topics
Mitigation options and costs
Innovation
Uncertainty
International policy architectures and coalition theory
Optimal balance between mitigation and adaptation
32
WITCH Model, Description and Applications
Uncertainty
Major Areas of Research:
 Stochastic WITCH
 Analysis of optimal investment trajectories under uncertainty
 Uncertainty on R&D productivity
 Policy uncertainty
Key Findings:
 Modeling innovation in a backstop technology as an uncertain process
leads to higher optimal levels of R&D investments
 Uncertainty on the stringency of the mitigation target leads to high
mitigation activity if a stringent target has the chance to come into force
References:
 Bosetti, V. and M. Tavoni (2009), "Uncertain R&D, backstop technology and
GHGs stabilization", Energy Economics, 31(1): S18-S26.
 Bosetti, V., C. Carraro, A. Sgobbi, and M.Tavoni (2009) "Delayed Action and
Uncertain Targets. How Much Will Climate Policy Cost?" Climatic Change,
Forthcoming
33
WITCH Model, Description and Applications
Major Research Topics
Mitigation options and costs
Innovation
Uncertainty
International policy architectures and coalition theory
Optimal balance between mitigation and adaptation
34
WITCH Model, Description and Applications
International Policy Architectures - 1
Major Areas of Research:
 International climate policy architectures (Harvard Project on
International Climate Agreements)
 Stabilization costs, investments and innovation with different
degrees of cooperation
 Delayed participation of developing countries
 Optimal climate policy of high income countries in face of
delayed participation from low income countries
 The incentives to participate in and the stability of climate
coalitions
35
WITCH Model, Description and Applications
International Policy Architectures - 2
• Global coalition with CAT and transfers
• Global coalition with carbon tax recycled domestically
• Global coalition with REDD
• Climate Clubs (sub-coalitions)
• Dynamic coalitions: incremental participation based on
 Burden sharing rules
 Graduation
 Dynamic targets
• R&D and Technology coalition
36
WITCH Model, Description and Applications
International Policy Architectures - 3
Key feature
Name
CAT with
redistribution
Benchmark cap and trade
Global carbon tax
Global tax recycled domestically Carbon Tax
REDD
Inclusion of REDD
Climate Clubs
Scope
Timing
Universal
Immediate
Universal
Immediate
Cap and Trade
Universal
Immediate
Clubs of countries
Cap and Trade
and R&D
Partial
Incremental
Burden Sharing
Delayed participation of DCs.
Cap and Trade
Universal
Incremental
Graduation
Bottom up targets
Cap and Trade
Partial
Incremental
Dynamic Targets
Political feasibility
Cap and Trade
Universal
Incremental
R&D Coalition
R&D cooperation
R&D
Universal
Immediate
NB All refer to CO2 only
37
Policy
Instrument
WITCH Model, Description and Applications
Cap and Trade
Climate Effectiveness
°C above pre-industrial
4.00
3.50
3.00
2.50
2.00
1.50
1.00
0.50
0.00
U
BA
C
AT
w
ith
r
b
tri
s
i
ed
io
ut
n
C
e
at
lim
ub
Cl
s
R
ED
D
rd
Bu
en
S
rin
a
h
g
ra
G
tio
a
du
n
l
G
a
ob
l
rb
ca
on
x
ta
D
am
n
y
ic
Ta
et
g
r
s
R
&D
C
o
iti
al
on
None of the policy architectures is able to keep temperature change below the 2°C
threshold. A target between 2.5 and 3°C seems more feasible
38
WITCH Model, Description and Applications
Economic Efficiency
R&D Coalition
Dynamic
Targets
Global carbon
tax
Graduation
Burden
Sharing
REDD
Climate Clubs
CAT with
redistribution
Change in GWP wrt BaU - Discounted at 5%
0.50%
0.00%
-0.50%
-1.00%
-1.50%
-2.00%
While temperature change varies less across the eight architectures for agreement
because of the inertia in the climate system, the economic costs of the different setups vary considerably. More stringent policy architectures imply a higher GWP loss
39
WITCH Model, Description and Applications
Non-Cooperative CO2 Emissions
The non-cooperative solution, defined also as the baseline, it best represents
the strategic nature of international relations. Little variations are observed in a
non-cooperative setting, reflecting the inability of individual regions to
internalise the environmental externality
40
WITCH Model, Description and Applications
Cooperative CO2 Emissions
Sensitivity to these assumptions is far greater in the cooperative case. Higher
damage and especially low discounting drive emissions down
41
WITCH Model, Description and Applications
CBA: Free riding – the case of SSA
USA
Percentage Difference in Emissions
Coalition w/o SSa comapred to the Grand Coalition
14%
WEURO
300%
EEURO
AUCANZ
12%
250%
JPNKOR
TE
10%
200%
8%
MENA
SASIA
150%
CHINA
6%
100%
4%
SEASIA
LAM
50%
2%
0%
SSA (RH Axes)
0%
2005
2010
2015
2020
2025
2030
2035
2040
2045
2050
When Africa leaves the grand coalition
• members emit more because they do not internalize the high negative impact of
climate change on Africa (damage effect)
• Africa emits more (free riding effect), but less than in the BaU (technology spillovers)
42
WITCH Model, Description and Applications
International Policy Architectures - 4
Major Findings:
 Delayed and fragmented participation of developing countries into
international climate agreements would raise the global policy costs
considerably for serious stabilization targets
 An international carbon market has the potential to alleviate such
detrimental effects, but might involve large financial transfers
 An agreement that envisions future commitments for some key
emerging economies might represent a win-win strategy, since the
optimal investment behavior is to anticipate climate policy
 This is especially relevant for China, whose recent and foreseeable
trends of investments in innovation are not incompatible with the
adoption of domestic emission reduction obligations in 2030
 In cost-benefit setting, only the Grand Coalition finds profitable to
achieve the 550 ppm CO2-eq target, under very special condition
 The Grand Coalition is neither stable nor potentially stable
43
WITCH Model, Description and Applications
International Policy Architectures - 5
References:
 Bosetti, V., C. Carraro, E. De Cian, R. Duval, E. Massetti and M. Tavoni (2009),
"The Incentives to Participate in and the Stability of International Climate
Coalitions: a Game Theoretic Approach Using the WITCH Model," OECD
Economics Department Working Papers No. 702, June 2009.
 Bosetti, V., C. Carraro and M.Tavoni (2009), " Climate Policy After 2012.
Technology, Timing, Participation,” CESifo Economic Studies, Forthcoming.
 Bosetti, V., C. Carraro and M.Tavoni (2009) " Climate Change Mitigation
Strategies in Fast-Growing Countries: The Benefits of Early Action”, Energy
Economics, Forthcoming.
 Bosetti, V., C. Carraro, A. Sgobbi, and M. Tavoni (2008). “Modelling Economic
Impacts of Alternative International Climate Policy Architectures: A Quantitative
and Comparative Assessment of Architectures for Agreement”, in Aldy and
Stavins, eds, Post-Kyoto International Climate Policy: Implementing
Architectures for Agreement Cambridge University Press, in press.
 Bosetti, V., C. Carraro and M. Tavoni (2008), "Delayed Participation of
Developing Countries to Climate Agreements: Should Action in the EU and US
be Postponed?", FEEM Working Paper N.70-2008.
44
WITCH Model, Description and Applications
Major Research Topics
Mitigation options and costs
Innovation
Uncertainty
International policy architectures and coalition theory
Optimal balance between mitigation and adaptation
45
WITCH Model, Description and Applications
Balancing Mitigation and Adaptation Policies
Major Areas of Research:
 Optimal mix of mitigation and adaptation policies
 Optimal investments in different adaptation forms
Key Findings:
 The introduction of adaptation decreases the need to mitigate and vice-versa
 Joint implementation of mitigation and adaptation in a cost-benefit framework
suggests that both policies are required
 Proactive adaptation is the first measure to be adopted. Reactive measures
prevail afterwards, when the damage is higher, and in non-OECD regions
 Developed countries are likely to experience minor aggregate
damages/benefits. Policy to control damages should focus on developing
countries
References:
 Bosello, F., C. Carraro and E. De Cian (2009). “An Analysis of Adaptation as a
Response to Climate Change.” Copenhagen Consensus Center, July 2009
 Bosello, F., C. Carraro and E. De Cian (2009). “Adaptation, Mitigation and Innovation: A
Comprehensive Approach to Climate Policy .” Fondazione Eni Enrico Mattei,
September 2009, mimeo
46
WITCH Model, Description and Applications