1 of 27

Assessing transboundary and intersectoral spillovers of multiple natural hazards in the Danube Region using a large-scale macroeconomic agent-based model

Nikita Strelkovskii, Sebastian Poledna, Stefan Hochrainer-Stigler,

Robert Šakić Trogrlić, Karina Reiter, Benjamin Blanz, James Daniell

The MYRIAD-EU project has received funding from the European Union’s Horizon 2020 research and innovation programme call H2020-LC-CLA-2018-2019-2020 under grant agreement number 101003276

@Myriad_EU

2 of 27

Trade dependencies in the Danube region

  • Germany is a central trading partner for all Danube countries, the import dependencies are generally lower in percentage terms than export dependencies
  • Hungary and Slovakia have significant mutual trade dependencies
  • Croatia has relatively lower trade dependencies with other Danube countries compared to the rest

​

@Myriad_EU

3 of 27

Economic spillover effects

  • Economies are deeply linked, and disruptions in one area can have widespread consequences
  • Economic spillover effects of natural hazards extend beyond direct damages, affecting supply chains, labor markets, financial stability, and public sector resources
  • Understanding economic spillovers can help in designing better disaster preparedness and response strategies

@Myriad_EU

4 of 27

Research question

How can we increase resilience to multiple disasters that impact several interconnected countries in the Danube Region with strong macroeconomic relationships?

https://danube-region.eu

@Myriad_EU

5 of 27

Damage mechanisms

  • Floods and earthquakes damage capital (e.g., destroyed buildings, infrastructure, and machinery)
  • Droughts directly damage output (e.g., reduced agricultural yields, water supply, and hydropower production)

​

@Myriad_EU

6 of 27

ABM version setup

  • 26 EU countries (except MT); 237 NUTS-2 regions
    • Including 9 Danube Region countries
  • 62 industries (NACE 2-digit)
  • Calibrated with FIGARO IO tables and Eurostat data
  • 3-year time horizon, quarterly timestep
  • 1:100 scale => ~300k firms and ~4,4M people

@Myriad_EU

7 of 27

Modeling indirect losses with the ABM

Coupling ABM with a damage scenario (Bachner et al. 2023)

Indirect risks arise not from the direct impact of a disaster but from the interconnectedness of system elements. They materialize through disruptions in transport, supply chains, or economic activities.

@Myriad_EU

8 of 27

Modeling indirect losses with the ABM

Coupling ABM with a damage scenario (Bachner et al. 2023)

@Myriad_EU

9 of 27

Capital losses implementation

  •  

@Myriad_EU

10 of 27

Capital losses implementation

  •  

@Myriad_EU

11 of 27

Investment response to capital losses

  •  

@Myriad_EU

12 of 27

Investment response to capital losses

  •  

@Myriad_EU

13 of 27

Credit constraints

  •  

@Myriad_EU

14 of 27

Output losses implementation

  •  

@Myriad_EU

15 of 27

Capital damage scenarios: flood and earthquake

@Myriad_EU

16 of 27

Capital damage scenarios: flood and earthquake

  • Floods most severely directly affect the agricultural industry
  • Earthquakes most severely directly affect trade and hospitality industries (but also almost all others)

@Myriad_EU

17 of 27

Output damage scenario: drought

  • Droughts most severely directly affect agricultural, water supply, and energy industries

​

@Myriad_EU

18 of 27

Multi-hazard scenarios

  • Both single- and multi-hazard scenarios, i.e., consecutive (a flood is followed by a drought in several countries) or compound (an earthquake in one country and a flood in another happen in one quarter) are implemented
  • Direct damage values are based on past events (i.e., 2010 Danube flood, 1940 earthquake in Romania, 2018 drought)

@Myriad_EU

19 of 27

Results: total output

  • Single hazards generally reduce the total output compared to the baseline
  • Consecutive hazards have a more severe impact, leading to lower total outputs
  • The total output often recovers over time, but does not usually reach the baseline level
  • The recovery rate can vary significantly between countries and scenarios

@Myriad_EU

20 of 27

Results: Unemployment

  • Countries slowly recover from the initial (smaller) shocks; a more severe shock discards this progress
  • Some countries, even those not hit by shocks at all, face secondary effects
  • In some countries, even those directly suffering from the shocks, unemployment levels decrease

​

@Myriad_EU

21 of 27

Results: understanding intersectoral and interregional spillover effects

  • Impact of the multi-hazard scenario on outputs of industries
  • Change in industrial outputs compared to the baseline

@Myriad_EU

22 of 27

Results: outputs of the agricultural and construction industries

@Myriad_EU

23 of 27

Changes in trade flows between the countries

@Myriad_EU

24 of 27

Takeaways

  • There is a strong dependency between countries during disaster events
    • Non-affected countries helping the affected countries by substituting the inputs
  • These dependencies are different according to time and sectors
    • Some non-affected industries & countries (e.g., the finance sector in Luxembourg) can be adversely affected – with a delay after the shock

​

@Myriad_EU

25 of 27

Current limitations

  • The current implementation of credit constraints does not completely reflect reality
  • Goods and services can be exported & imported in case of a shock, even if unfeasible (common market for all countries)
    • Realistic supply chains can help

​

@Myriad_EU

26 of 27

Thank you for your time!��Questions?

The MYRIAD-EU project has received funding from the European Union’s Horizon 2020 research and innovation programme call H2020-LC-CLA-2018-2019-2020 under grant agreement number 101003276

@Myriad_EU

27 of 27

myriadproject.eu

The MYRIAD-EU project has received funding from the European Union’s Horizon 2020 research and innovation programme call H2020-LC-CLA-2018-2019-2020 under grant agreement number 101003276

@Myriad_EU