External costs in the MARKAL framework

Externalities in the bottom-up

energy system modeling framework Analyses with the MARKAL model

Peter Rafaj

International Institute for Applied Systems Analysis (IIASA)

Expert Workshop on Energy and Climate Change Modeling 17th of November 2011 Seoul, Korea

Scope

Definition of externalities

Methodology for internalisation of external costs

Scenario results & sensitivities

Combining externalities with other policies

Insights from modelling experiments

Externalities and energy system

External costs are introduced if

• the emissions from the energy system imply damages to the society

• the resulting costs are not included in the market price of energy

Internalisation of external costs intends to

• compensate for the health and environmental damages

• yield a full-cost pricing of energy services

Beside the air emissions, additional externality burdens are considered:

• solid and liquid wastes, risk of accidents, occupational exposure to hazardous substances, noise, others

Quantification and monetization of damages requires

• site-specific impact assessment of technologies

• comparisons between different energy chains and fuel cycles

External costs in the MARKAL framework

Different methods applicable:

1. Ex-post quantification of damages and valuation of impacts - no feedback into the optimisation

2. Externality charged to every unit of output

3. Damage function implying a tax on air emissions

 ^

^





t

t t

extern

Z ExtCost ypp Q d

Z * * * ( 1 )

¹





poll

poll r t poll

r t r

t

DV EM

DAM

_{, , ,}

*

_{, ,} ^

Ex Post Analysis of Externalities using MARKAL

Impact pathway approach

MARKAL India MARKAL Pakistan

GAINS ASIA

Greenhouse gas–Air pollution

Interactions and Synergies model

•Emission sources Energy

scenarios •Dispersion

•Exposure Emission

scenarios

•Health

•Ecosystems

•Crops Dose -

response functions

Externality valuation

ALPHA

Atmospheric Long-range Pollution Health-environment Assessment

ExternE, NEEDS Soft link interfaces

Monetization of

damages (€million/year)

Economic Benefits of Climate Mitigation Policy

Example India

0 25 50 75

2000 2010 2020 2030

EJ/yr

Nuclear Renewables Biomass Gas Oil Coal

0 25 50 75

2000 2010 2020 2030

EJ/yr

Nuclear Renewables Biomass Gas Oil Coal Baseline

Low carbon

Baseline (2030)

Low carbon (2030) Energy projections: MARKAL Emissions and health impacts: GAINS

Baseline 2030

Ozone mortality 377 Ozone morbidity 572 PM2 .5 mortality 227,442 PM_{2 .5} morbidity 86,655

Total 315,046

Low carbon 2030 Ozone mortality 304 Ozone morbidity 461 PM2 .5 mortality 185,314 PM2 .5 morbidity 70,605

Total 256,684

Co-benefit 58,362

Regions: _NAME

OOECD

EEFSU ^{LAFM ASIA}

Integration in the Global Markal Model (GMM)

Main features

“Bottom-up” techno-economic model → Explicit representation of technologies

Optimisation under perfect foresight assumptions Time horizon 2000-2050, 10-year steps

Partial equilibrium → Elastic demands Energy system of five world regions →

Multi-regional trading of selected commodities Endogenous technological learning

Learning spill-over across regions

Internalisation of externalities in power sector

Basic assumptions

• External costs from local pollution (SO₂, NO_x, PM) and/or CO₂ internalized in the power sector

• External costs for each power plant in ¢ /kWh derived from the EU ExternE-Project

• Externalities adjusted for regional differences in population density, fuel quality, power-plant efficiency and application of emission-

control systems

Determinant for scaling Unit SO₂ NO_x PM CO₂

Average damage cost per pollutant €1995/t 8000 7000 14000 19

High 1.5 1.5 1.5

Medium 1 1 1

Population density adjustment factor (AF)

Low 0.75 0.75 0.75

n.a.

coal oil natural gas Reference thermal efficiency

% 41 40 55

Scaling of external costs

Options considered

1. Population density

• NAME, EEFSU, LAFM – Medium

• ASIA, OOECD - High

2. Improved conversion efficiency

3. Welfare in regions (GDP/cap)

t t original t

original t ExtCost

ExtCost



 _{, 0}

0

, _ * ^



EU t ppp

r t ppp r

t original r

t GDP

ExtCost GDP ExtCost

0 ,

, ,

,

, *





EU t mex

r t mex r

t original r

t GDP

ExtCost GDP ExtCost

0 ,

, ,

,

, *





Fossil-fuel based power plants min max min max

Coal conventional 8.1 19.0 9.8 20.8

Coal conventional with DeSO_x/DeNO_x 1.2 1.8 2.9 3.6 Coal conv. with DeSO_x/DeNO_x and CO₂ seq 1.5 2.3 1.8 2.9

Coal advanced 1.6 2.4 2.8 3.8

Coal advanced with CO₂ seq 1.8 2.8 1.9 3.0

Coal IGCC 0.5 1.0 2.2 2.9

Coal IGCC with CO₂ seq 0.6 1.2 1.2 1.7

Natural Gas Combined Cycle (NGCC) 0.3 1.1 0.8 1.7 NGCC with CO₂ sequestration 0.3 1.3 0.7 1.5 Gas steam conventional 1.1 3.0 1.9 3.8 Cogenaration gas turbine 1.2 2.3 2.2 3.3

Oil conventional 1.3 5.9 2.5 7.2

Non-fossil power plants

Nuclear plant - Light Water Reactor (LWR) Hydro-electric plant (small and large) Solar photovoltaics (SPV)

Wind turbine

Biomass power plant

Geothermal electric 0.1

0.3 0.1 0.1 0.1

0.4 0.4 0.1 0.3 0.1 0.5 External cost (cent/kWh)

0.5 Technology

excl CO₂ incl CO₂

Total electricity generation cost analysis

Example Asia, year 2050

0 20 40 60 80 100 120 140 160 180 200

Coal conv.

Coal conv. DENOx/DESOx Advanced coal

Advanced coal CO2 seq IG CC

IG CC C

O2 seq

NGCC NGCC C

O2 seq

Gas steam Gas fu

el cell

Nuclear LWR Nuclear advanced

Hydropower

SPV Wind tu

rb ine

H2 Fuel cell

Electricity generation cost (mill $/kWh) Local&Global

external cost Local external cost

Fuel cost C-storage cost O&M cost Investment cost Generation cost 2000

Q E Q F Q

M VARO

Q M FIXO Q

CRF

TGC I *  &  &  

Development in global electricity production

Fuel mix changes due to integration of externalities

0 10000 20000 30000 40000 50000 60000 70000

Baseline Baseline Local externality Global externality Baseline Local externality Global externality Baseline Local externality Global externality

Power generation by fuel (TWh/yr)

2000 2020 2030 2050

Renewables Hydro Nuclear Coal Oil

Natural gas

Impact on electricity generation profile

Technology portfolio in 2050

0 5000 10000 15000 20000

NGCC NGCC CO2 seq Gas conv Gas FC Oil Coal conv Coal DeSOx/DeNOx Coal adv Coal adv. CO2 seq IGCC IGCC CO2 seq LWR NNU Hydro SPV Solar thermal Wind Biomass Geothermal Hydrogen FC

Power generation mix in 2050 (TWh/yr)

Baseline

Local externality Global externality

Global air emissions

CO₂ from all sources; SO₂/NO_x from power sector

0 10 20

2000 2020 2030 2050 GtC/yr

CO₂

0 50 100

2000 2020 2030 2050 MtSO2/yr

SO₂

0 25 50

2000 2020 2030 2050 MtNOx/yr

NO_x

Baseline

Local externality Global externality Scaled to GDP/cap

• Alternative scaling of externalities with GDP results in lower cost penalty, still the impact on emissions is significant.

CO

₂

emissions reduction components

Relative to the Baseline

0 20 40 60 80 100

2010 2020 2030 2040 2050

Contribution to CO2-reduction(%)

Local externality

CO2-capture

Nuclear fraction increase

Renewable fraction increase

Inter-fossil fuel switch

Demand reductions 0

20 40 60 80 100

2010 2020 2030 2040 2050

(%)

Global externality

CO2-capture

Nuclear fraction increase

Renewable fraction increase

Inter-fossil fuel switch

Demand reductions

Change in the cumulative energy system cost,

including external cost fraction

19.8

1.7 3.0

16.0

1.7

11.9

0.0 1.1

23.5

10.3

13.2

19.8

9.6

14.8

7.5

0 5 10 15 20 25

Baseline Non-

internalized external cost

in Baseline

Local externality

Global externality

Non- internalized external cost

in Baseline

Local externality

Non- internalized external cost

in Baseline

Local externality

Change in total systemcost vs. Baseline(%)

NON-INTERNALISED GLOBAL EXTERNALITY

NON-INTERNALISED LOCAL EXTERNALITY

EXTERNALITY FRACTION

TECHNOLOGY CHANGE AND FUEL

SUBSTITUTION

Scaled to population density Scaled to GDP_PPP/cap Scaled to GDP_MEX/cap

Synergies in combined policy adoption

Global CO₂ emission reductions

-50 -40 -30 -20 -10 0

2000 2010 2020 2030 2040 2050

Reduction in CO2 emissions over Baseline

CO2-cap&trade Renewable portfolio Local externality

CO2-cap&trade + Local externality Renewable portfolio + Local externality

%

Single policies

Cost and Benefit Assessment L

arge uncertainties

-4.3

-3.0

-8.8

-5.3

-9.0

-15 -12.5 -10 -7.5 -5 -2.5 0 2.5 5

CO₂-Cap&Trade

Renewable Portfolio

Local Externality

CO₂-Cap&Trade + Renewable Portfolio

CO₂-Cap&Trade + Local Externality

(%)

Increase in total energy system cost

Ancillary energy system co-benefits

Avoided damages from air pollution (scaled to GDPppp/cap)

Policy cost Secondary benefits

Relative to Baseline

Conclusions

• Monetary evaluation of the (co)benefits of emission control strategies provides relevant insights for decision makers

• Quantification of impacts based on MARKAL-inputs, but outside the optimization procedure, brings detailed assessment of a policy, when linked with dedicated air quality models (GAINS)

• Externalities integrated in the MARKAL’s cost function allows to balance trade-offs between environmental ambition and the economic implications

• Modeling results indicate a large scope of co-benefits resulting from the parallel application of different policy instruments

• Monetization of health & environmental benefits are associated with a wide range of uncertainties and controversies

• If the externality analyses are used in an international policy context, it is challenging to attribute economic values to non-market goods: human life and ecosystems