IIASA perspective on socio-economic disaster impact modelling
Dr. Reinhard Mechler& Dr. Stefan Hochrainer-Stigler
SCIENTIFIC SEMINAR ON NATURAL DISASTERS: BRIDGING SCIENCE-BASED EARLY WARNING AND EARLY ACTION
DECISION MAKING
7-8 November 2012, Brussels, Centre Albert Borschette
4 issues
1. Assessment of socio-economic impacts moving towards risk-based assessment of economic impacts, but uncertainties large
2. Increasing emphasis of vulnerability/adaptive capacity
3. Cascading/multi-hazard assessments being worked on
4. Making the economic case for DRM challenged by lack of robustness
Assessment framework: Direct and indirect risk
Impacts on Assets Impacts on FlowsRisk as a function ofExposure, Hazard, Vulnerability
Costs of extremesIssues
• Direct damage costs fairly well studied (in OECD countries)• Little known about intangibles• Indirect costs: are important, but less is known• Recently, risk based approaches taking a probabilistic and
forward looking approach emerging
Exposure: population and capital stock exposed Intensity vs. damage
Intensity and frequency
Climate change
Adaptation
Stronger focus on risk in disaster impact modelling as well as climate adaptation analyses
Risk triangle
Efficiency of risk management instruments depends on occurrence probability
EU solidarity fund
Market based insurance
Flood protection
Why Risk?
EU impact costs assessments for weather related risks
• EU research projects (MICE, ESPON, MARS, ENSEMBLES, PESETA, ADAM, ClimateCost) have focussed on weather-related hazards and risks in Europe.
• National-level assessments of current and future weather risks, mostly on flood risk in England and Wales (DEFRA, 2001), Germany (Apel et al., 2004, and Merz and Thieken, 2004) and the United States (Scawthorn et al., 2006a, 2006b). Hall, Sayers and Dawson (2005) projected risks up to 2100 for England and Wales.
• PESETA project fills an important gap by conducting a European-wide assessment of current and future flood risks up to 2100 and estimated average annual values.
• IIASA based on ADAM flood and drought risk at a European scale identifying monetary economic losses, but estimated full probability distributions at different aggregation scales (see Mechler et al., 2010)
0.98
0.985
0.99
0.995
1
0 10 20 30 40 50 60 70 80Billion Euro
F
Austria Belgium Bulgaria Czech Republic Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Netherlands Poland Portugal Romania Slovakia Slovenia Spain Sweden United Kingdom
Towards probabilistic economic risk assessmentflood risk
Source: Hochrainer et al., 2011 based on Moriondo et al., 2010
IPCC-SREX:A changing climate leads to changes in extremeweather and climate events
Extreme event impact and risk assessments increasinglyimportant for climate adaptation assessments
Approach Description Examples Advantages Issues Economic Integrated Assessment Models (IAM)
Aggregated economic models. Values in future periods, expressed £ and %GDP and values over time (PVs)
Global studies (e.g. de Bruin et al) that provide outputs for Europe.
Provide headline values for raising awareness. Very flexible – wide range of potential outputs (future years, PV, CBA).
Aggregated and low representation of impacts, generally exclude extreme events and do not capture adaptation in any realistic form. Not suitable for detailed national planning.
Investment and Financial Flows (I&FF)
Financial analysis. Costs of adaptation (increase against future baseline)
Global studies (e.g. UNFCCC, 2007). National studies, e.g. Swedish Commission (2007) has analysis with I&FF type approach.
Costs of adaptation in short-term policy time-scale. Easier to apply even without detailed analysis of climate change.
No specific linkage with climate change or adaptation (though can be included). No analysis of adaptation benefits or residual impacts.
Computable General Equilibrium models (GCE)
Multi-sectoral economic analysis.
National level – Germany (Kemfert, 2006)
Capture cross-sectoral linkages in economy wide models (not in other approaches). Can represent global and trade effects.
Aggregated representation of impacts and can only capture adaptation in market form. Issues with projections of sectoral linkages. Omits non-market effects. Not suitable for detailed national planning.
Impact assessment (scenario based assessment)
Physical effects and economic costs of CC with sectoral models in future periods, and costs and benefits of adaptation or in cost-effectiveness analysis
PESETA study (2007) and coastal analysis. National scale: UK Flooding (Thorne et. al. 2007)
More sector specific analysis. Provides physical impacts as well as economic values – therefore can capture gaps and non-market sectors.
Not able to represent cross-sectoral, economy-wide effects. Tends to treat adaptation as a menu of hard (technical) adaptation options. Less relevant for short-term policy.
Impact assessment - shocks
Use of historic damage loss relationships (statistics and econometrics) applied to future projections of shocks combined with adaptation costs (and sometimes benefits)
Sector level, e.g. EAC study (2009) in the UK.
Allow consideration of future climate variability (in addition to future trends)
Issues of applying historical relationships to the future. Issues with high uncertainty in predicting future extremes.
Impact assessment - econometric based
Relationships between economic production and climate parameters derived with econometric analysis and applied to future scenarios – and to consider adaptation
National level Household level or sector
Can provide information on overall economic growth and allow analysis of longer-term effects. Provide greater sophistication with level of detail.
Mostly focused on autonomous or non-specified adaptation. Very simplistic relationships to represent complex parameters. No information on specific attributes. Issues on whether relationships are applicable to future time periods.
Risk management Current and future risks to climate variability. Probabilistic approach.
Flood risk studies (coastal and river).
Well suited for current and future risks and uncertainty, Often used with Cost-effectiveness. Has been applied in adaptive management and iterative analysis.
Extra dimension of complexity associated with probabilistic approach. Limited applicability: focused on thresholds (e.g. risk of flooding).
Adaptation assessments
Risks over a range of policy / planning horizons. Often linked risk management and adaptive capacity.
No real economic examples. Emerging number of adaptation assessments.
Stronger focus on immediate adaptation policy needs and decision making under uncertainty and greater consideration of diversity of adaptation (including soft options) and adaptive capacity.
Less explored in relation to economic assessment
Source: Watkiss and Hunt, 2010
Impact assess-ment
Risk management
But uncertainties are large even today Average annual flood losses (AAL)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4A
T
BE
BG CZ
DK
EE FI FR D
GR
HU IE IT LV LT LU NL
PL
PT
RO
SK
SE
ES S
UK
min max AAL
Billi
on E
uro
Annual average losses including confidence bounds for minimum and maximum estimates. Source: Based on Hochrainer and Mechler, 2009 and Lugeri et al. 2010
Flooding: risk and uninsured risk
Total annual average flood risk: 3.4 billion Euro
AT B BU CZDK
ESFIF
D
G
HIRITLVLTLUNLPLPOROSK
SLESS
UK
Total annual uninsured average flood risk: 2.3 billion Euro
AT B BUCZDK
ESFI
FD
GHIRITLV
LTLUNLPL
POROSK
SLESS
UK
Distribution of flood risk including the uninsured risk over Europe in absolute terms Note: the uninsured risk comprises private and public assets, and the latter are generally uninsu
Source: IIASA, 2011 for DG CLIMA study
Bräuninger, M., Butzengeiger-Geyer, S., Dlugolecki, A., Hochrainer, S., Köhler,K., Linnerooth-Bayer, J., Mechler, R., Michaelowa, A.,Schulze, S. (2011). Application of economic instruments for adaptation to climate change. Report to the European Commission, Directorate General CLIMA, Brussels.
Windstorm risk (hotspot countries and 200 year event only)
Distribution of windstorm risk for the 200 year event including the uninsured risk over Europe in absolute terms
Total risk: 60.4 billion Euro
AT BE DK
F
DIELUNL
UK
Uninsured risk: 22.4 billion Euro
AT BE DK
F
DIELUNL
UK
Source: IIASA, 2011 for DG CLIMA study
Bräuninger, M., Butzengeiger-Geyer, S., Dlugolecki, A., Hochrainer, S., Köhler,K., Linnerooth-Bayer, J., Mechler, R., Michaelowa, A.,Schulze, S. (2011). Application of economic instruments for adaptation to climate change. Report to the European Commission, Directorate General CLIMA, Brussels.
Drought risk
Distribution of drought risk including the uninsured risk over Europe in absolute terms
Source: IIASA, 2011 for DG CLIMA study
Bräuninger, M., Butzengeiger-Geyer, S., Dlugolecki, A., Hochrainer, S., Köhler,K., Linnerooth-Bayer, J., Mechler, R., Michaelowa, A.,Schulze, S. (2011). Application of economic instruments for adaptation to climate change. Report to the European Commission, Directorate General CLIMA, Brussels.
Modelling indirect risksKey transmission mechanism is vulnerability
Source: Kohler et al., 2004
Physical adaptation (hard…) Socio-economic adaptation
(soft…) incl. focus on economic instruments
Increasing emphasis onenhancing “soft-
resilience” shaped by socioeconomic factor vs.
hard resilience
Example Estimating fiscal vulnerability
Source: IIASA CATSIM model
Government fiscal deficits and hidden liabilities due to flood risk in selection of flood prone European countries
Source: Mechler et al.,2010
0%
1%
2%
3%
4%
5%
6%
Aus
tria
Hun
gary
Rom
ania
Cze
chR
epub
lic
Latv
ia
Pol
and
Bulg
aria
Slov
akia
Lith
uani
a
Per c
ent o
f GD
P
Government flood risk liabilityProjected fiscal deficit 2009
0%
1%
2%
3%
4%
5%
6%
Aus
tria
Hun
gary
Rom
ania
Cze
chR
epub
lic
Latv
ia
Pol
and
Bulg
aria
Slov
akia
Lith
uani
a
Per c
ent o
f GD
P
Government flood risk liabilityProjected fiscal deficit 2009
Hidden government disaster liabilities can be large
IIASA developing a methodology for assessing multiple hazards and cascading risks: MATRIX project
Disaster events can be triggered by multiple hazards and lead to cascading impacts across regions
Tōhuku event 2011
Making the economic case for DRM
• A lot of rhetoric regarding high economic returns to DRM, yet little solid evidence• Generally: BC ratios may be up to 4• But IPCC-SREX: “Rigorous CBA for managing extreme events seems limited based on
limited evidence and medium agreement in the literature.”• Other methods for decision-support of interest: MCA, robust approaches
Source: Mechler, 2012
Thank you for the attention, and apologies for not being able to attend!
Contact:[email protected]@iiasa.ac.at
