T6 – Towards a pan-European drought risk insurance pool: From impact modelling to insurance design

Short description

As climate change intensifies hydro-meteorological hazards, the economic impacts of droughts are expected to increase in Europe [1]. Already under current conditions, EU-wide estimates indicate that catastrophic agricultural losses (2% annual probability) can reach about €35.1 billion. Yet drought remains highly underinsured across much of Europe, with only 20-30% of climate-related agricultural losses being insured [2]. Current drought risk management and insurance are also fragmented across Member States, with low penetration rates and strong reliance on post-disaster public aid. This fragmentation raises concerns about the sustainability, fairness, and efficiency of drought protection in a changing climate, and has stimulated interest in more coordinated EU-level approaches. From an insurance perspective, pooling and diversification are central to strengthening insurance adoption and affordability: bundled portfolios with less correlated risks require lower reserves and can reduce premiums, motivating designs that pool risk across space, sectors, and hazards [3].

This thesis will develop a modelling framework to evaluate whether a pan-European drought insurance pool can increase coverage while improving affordability and equity. Potential complementary strategies could also be assessed, including hazard bundling (e.g., drought and flood), coordinated risk-reduction investments (e.g., incentives in irrigation systems improvements), and alternative parametric insurance schemes like collar or hybrid contracts [4]. First, the thesis will build data-driven impact models for drought-related losses for three economic sectors: (i) agriculture, (ii) navigation, and (iii) hydropower. For agriculture (main sector considered), these models will be calibrated and tested across Europe using as predictors hydro-meteorological data and drought indices, including Standardized Precipitation Index (SPI), Standardized Precipitation-Evapotranspiration Index (SPEI) at multiple aggregation scales, and Low Flow Index (LFI), from the European Drought Observatory (EDO) of the Copernicus Emergency Management Service (CEMS). Crop yield data (losses) at the annual scale will be used as the primary target. Second, the thesis will extend an existing insurance-reform model (currently available for floods) to drought and simulate alternative market designs: fully risk-based, solidaristic, and public–private partnership schemes. These market configurations will be tested under (i) national schemes and (ii) a single EU-wide pool. Finally, the framework will assess how these insurance designs interact with proactive public investments in drought risk reduction at the national or EU level, and quantify possible trade-offs among coverage, total costs, and cross-country equity. Outputs will include: (i) estimates of historical/current (and, where feasible, scenario-based) drought impacts for agriculture and possibly energy and navigation; (ii) an evaluation of different insurance market designs and the associated protection gap, under national-based, uncoordinated schemes; (iii) a quantitative assessment of how EU-level risk pooling (across countries, sectors, and potentially hazards) and risk reduction investments could improve insurance performance and equity at lower overall cost.

PROPOSED ACTIVITIES

Literature Review

  • Conduct a systematic literature review covering drought risk, impact modelling, insurance, and financial risk pooling
  • Review EU-wide evidence (reports and grey literature) on agricultural climate risk and insurance systems, including risk metrics used by practitioners, e.g., Probable Maximum Loss (PML), Annual Average Loss (AAL)

Data Processing and Analysis

  • Extract and analyse several drought indices over Europe, including the Standardized Precipitation Index (SPI), Standardized Precipitation-Evapotranspiration Index (SPEI), the Low Flow Index (LFI), from the CEMS European Drought Observatory (EDO)
  • Develop and validate sectoral impact models, starting from simple regressions and complexifying as needed; produce pan-EU impact maps and time series

Insurance Design and Simulation Experiments

  • Apply and extend an existing insurance-reform model to drought risks and simulate alternative designs (market configurations and pool levels)
  • Evaluate insurance reforms and pool benefits, comparing designs across multiple objectives

References

  1. Naumann et al. (2021), “Increased economic drought impacts in Europe with anthropogenic warming”. Nat. Clim. Chang. 11, 485–491. https://doi.org/10.1038/s41558-021-01044-3
  2. Fi-compass (2025), “Insurance and Risk Management Tools for Agriculture in the EU”. https://www.fi-compass.eu/library/market-analysis/insurance-and-risk-management-tools-agriculture-eu
  3. Broberg and Hovani-Bue, (2019). “Disaster Risk Reduction through Risk Pooling – The Case of Hazard Risk Pooling Schemes”. https://doi.org/10.1017/9781108564540.015
  4. Scarpellini et al. (2025), “Parametric insurance for drought and market impacts mitigation in the hydropower sector”. Water Resources Research, 61, e2025WR040147. https://doi.org/10.1029/2025WR040147

Relevant courses and knowledge

Advanced Environmental Systems Analysis

Number of students

2, preferably starting during the spring semester

Requisites

The student should be comfortable with climate change analysis (impact assessment), statistical analysis (e.g., extreme values), and data handling. Proficient coding skills (Python or R) are mandatory.