This document summarizes a research project on climate risk management in Austria. It discusses (1) the background and goals of the RESPECT research project, which aims to develop integrated climate risk management concepts and tools in Austria with a focus on floods and droughts. It then summarizes (2) a stochastic debt assessment that models how flood risks may impact Austria's public finances and debt levels. Finally, it outlines (3) how participatory role-playing methods were used to support climate risk management at the local level in Lienz, Austria.

1. Multi-layered comprehensive climate
risk management (CRM) in Austria –
connecting local and national levels
Thomas Schinko, International Institute for
Applied Systems Analysis (IIASA)
Markus Leitner, Environment Agency Austria (EAA)
OECD High Level Risk Forum, 18-19/09/2019, Paris

2. Outline
1 Background & The RESPECT research project
2 A stochastic debt assessment of flood risk in Austria
3 Participatory methods for supporting local level CRM
2

3. Background
• Climate-related risks are increasing at global and national levels
• Multiple drivers of risk: hazard (climate variability & anthropogenic
climate change) and socioeconomics (exposure and vulnerability)
• How can disaster risk management (DRM) and climate change
adaptation (CCA) be better linked in practice?
• Integrated climate risk management (CRM) and multi-sectoral
partnerships
• Shared responsibilities: public - private
• “Multilevel Governance“ across different authorities
• „Policy-Implementation Gap“
• Differing priorities and planning horizons
• Unclear roles and responsibilities
• Cooperation mechanisms do not exist yet
3

4. Responsibility and Risk: Operationalizing comprehensive climate risk
layering in Austria among multiple actors

5. 0.51
2015)
-
5
10
15
20
25
30
35
40
2015 2030 2050
Losses(bnEuros2015)
100 250 500 1000
Risk
prevention
Insurance
Absorption
Monitoring
Climate risk
analysis
Evaluation
Implementation
Transformation
Continuation
Monitoring
Climate risk
analysis
Evaluation
Implementation
Transformation
Continuation
Iterative CRM and risk layering
Source: Schinko et al. (2016)

6. Goals and Content of RESPECT
• Goal: Developing concepts, methods and tools for supporting
comprehensive CRM in Austria (Focus on floods and
droughts)
• Content:
• Institutional and stakeholder analysis
• Quantitative risk assessment
• Identification of roles and responsibilities in CRM
• Two case-studies:
• National level: model-based assessment of fiscal flood risk
• Local level (Lienz): Role-play simulation for identifying roles
and responsibilities in local level CRM
6

7. Outline
1 Background & The RESPECT research project
2 A stochastic debt assessment of flood risk in Austria
3 Participatory methods for supporting local level CRM
7

8. Motivation
• 89 countries have adopted fiscal rules (IMF, 2015)
• Debt, budget balance, expenditure, revenue…
• EU monetary union: Stability and Growth Pact
• Annual government deficit < 3% GDP
• Debt-to-GDP ratio < 60% (Austria 2016: 83.6%)
• Demographic concerns considered a major driver
for fiscal pressure (EC, 2015)
• Ageing, unemployment & health care expend.
• Medium Term Budgetary Objective (MTO)
requires ‘front loading’ approach to
demographic contingent liabilities

9. Climate risk in public balance sheets
• Concerns over contingent climate-related public
costs have received little attention so far but
• Research shows that future climate-related fiscal
liabilities will not be negligible (e.g. for AT: APCC,
2014; Steininger et al., 2015; Schinko et al., 2016)
• 2014-2020 EU budget: at least 20% of the European
budget (Euro 1.7 billion) to be allocated for climate-
related expenses (EC 2013)
• Triannual longer term budget forecast for Austria
qualitatively highlights importance of climate risk
(BMF, 2016)

10. Background - Methodology
• Most modeling exercises have used non-
probabilistic approaches
• Potential consequences under “average”
conditions
• Little insight how societal trajectories might
deviate from average projections if extreme
events occur
• High uncertainties regarding climate and
socioeconomic development paths
•  probabilistic approaches

11. Aim and focus
• Aim
• Design a mainstreaming methodology to integrate
climate risk into longer-term fiscal planning and
governance
• Testing it by means of a stochastic debt model
• Focus
• Climate-related extreme events
• Public sector
• Case study for Austria
• Public costs of current & future riverine flood risk

12. Methodology – Mainstreaming
framework
Baseline Population and GDP
estimates (EUROPOP/SSPs)
HazardVulnerabilityExposure Baseline Climate Scenario
(RCPs)
Economic cost due to
climate extreme
Contingent liability due to
demography-related cost
Existing estimate of fiscal consolidation needs and fiscal
sustainability at EU level
Revised estimate of fiscal consolidation needs and fiscal
sustainability at EU level
Policy Assumptions
Other Macroeconomic & Fiscal
Assumptions
Source: Mochizuki et al. (2018)

13. Stochastic debt model
1
𝑑𝑡 = 𝑑𝑡−1
1+𝑖 𝑡
1+𝑔 𝑡
− 𝑏𝑡 + 𝑐𝑡 + 𝑗𝑡 + 𝑓𝑡 …(1)2
3
𝑑𝑡 = Debt to GDP ratio in year t4
𝑖𝑡 = Real implicit interest rate at year t5
𝑔𝑡 = Real GDP growth rate at year t6
𝑏𝑡 = Structural primary balance over GDP in year t7
𝑐𝑡 = Change in age-related costs over GDP in year t relative to base year8
𝑗𝑡 = Residual public contingent liability due to climate extreme events over GDP in year t9
𝑓𝑡 = Stock flow adjustment over GDP in year t10
11
𝑑𝑡 = 𝑑𝑡−1
1+𝑖 𝑡
1+𝑔 𝑡
− 𝑏𝑡 + 𝑐𝑡 + 𝑗𝑡 + 𝑓𝑡 …(1)
P ratio in year t
t interest rate at year t
rowth rate at year t
imary balance over GDP in year t
ge-related costs over GDP in year t relative to base year
blic contingent liability due to climate extreme events over GDP in year t
djustment over GDP in year t
…Stochastic variables

14. Stochastic scenarios
• Two types of stochastic shocks up to 2050
• Macroeconomic variability
• Monte-Carlo simulation of historical (2002-2015)
variance-covariance matrix of GDP & short-/long-run
interest rates (Berti, 2013)
• Flood damages (i.e. direct economic flood risk)
• Structured coupling of (LISFLOOD) loss distributions at
basin scale employing a Copula approach (e.g. Jongman
et al., 2014; Timonina et al., 2015)

15. Results: Baseline scenario SSP2
EC 2012 EC 2016 Present Study
Annual changes in primary balance needed to
stablize debt at 60% in 2030 (p.p. of GDP)
0.40a 0.30b 0.07c
Average annual changes in age-related
expenditured (p.p. of GDP)
0.09 0.08 0.19
Average annual flood losses 2015 (% of GDP) n.a. n.a. 0.10
Average annual flood losses 2030 (% of GDP) n.a. n.a. 0.12
Average annual flood losses 2050 (% of GDP) n.a. n.a. 0.14
100 year flood damage in 2015 (% of GDP) n.a. n.a. 2.80
100 year flood damage in 2030 (% of GDP) n.a. n.a. 3.30
100 year flood damage in 2050 (% of GDP) n.a. n.a. 3.80
Source: Mochizuki et al. (2018) based on EC (2012), EC(2016) and own estimation
Note: a constant adjustment needed for period 2014-2020 to stablize debt at 2030;b constant adjustment needed
for period 2018-2022 for stablization at 2030; cconstant adjustment needed for period 2015-2022 for
stablization at 2030. d excluding unemployment related costs.
Table 3. Fiscal Consolidation Needs, Ageing related Costs and Climate Extreme Costs

16. Results: Stochastic debt trajectories
Flood risk
Fig 4a: Stochastic debt trajectories for Austria under SSP2 scenario up to 2030, flood risk only.
Showing 5th to 95th percenties. Source: Mochizuki et al. (2018)

17. Results: Stochastic debt trajectories
Flood risk and macroeconomic variability
Fig 4b: Stochastic debt trajectories for Austria under SSP2 scenario up to 2030, flood risk and
macroeconomic variability. Showing 5th to 95th percenties. Source: Mochizuki et al. (2018)

18. Results: The Austrian Disaster fund
2015-2030 2031-2050
Probability of disaster fund
depletion
Under B/C ratio of 1:
15 %
Under B/C ratio of 4:
4.0%
Under B/C ratio of 1:
14%
Under B/C ratio of 4:
2.9%
Magnitude of fund depletion
(in million EUR 2015)
Under B/C ratio of 1:
Median: 280
SD: 1,750
Under B/C ratio of 4:
Median: 470
SD: 2,640
Under B/C ratio of 1:
Median: 380
SD: 2,780
Under B/C ratio of 4:
Median: 1,840
SD: 4,460
Table 4. Disaster Fund Simulation
Source: Mochizuki et al. (2018)

19. Discussion & Conclusions
• Expected flood damages small compared to macro-economic
variability and ageing costs
• Extreme event risk (100 year event) > annual changes in
age-related expenditure
• Flood risk alone unlikely to impact Austria’s fiscal position
• But: Current disaster fund arrangements not sustainable &
have to be reconsidered by allowing for
• Building back better; Private ex-ante risk reduction;
Streamlining with NatCat insurance; Public risk reduction
beyond physical measures; fat tail risks
• Requires climate risk mainstreaming
• E.g. within Climate Change Adaptation Strategies

20. Outline
1 Background & The RESPECT research project
2 A stochastic debt assessment of flood risk in Austria
3 Participatory methods for supporting local level CRM
23

21. Stakeholder Maps
• Separated for floods und water scarcity/drought
24

22. CRM Activities in Austria
• Stakeholder-Activity Matrix
• Separate for floods und water scarcity/drought
• Based on stakeholder mapping and CRM Cycle
25

23. Stakeholders’ recommendations to
operationalize CRM in Austria
• Institutionalization of a national climate risk council
supporting the diffusion of CRM in Austria
• Currently weak (ad-hoc) coordination between different areas
• Establishing decision making structures
• Periodically published climate risk report
• Separate or integrated (e.g. part of national security counsel)
• Extension of the Disaster Relief Fund act to support
preventive measures of the private sector
• Combination with private insurance schemes
• Implementation of a risk-aware policy needs a broad
support by the general population
• Importance of raising awareness for climate related risks
26

24. RESPECT Role Play
„Flood risk in Lienz“, June 2018
27
Foto-Credits: Michèle Lintschnig

25. 28
Thank you for your attention.
schinko@iiasa.ac.at and markus.leitner@umweltbundesamt.at
RESPECT project information and updates:
respectproject.net
Open Access publication in
Regional Environmental Change:
https://link.springer.com/article/10.1007/s10113-018-1300-3