How Is the Business Model of Insurers Affected by Weather?

What Can the Insurance Industry Contribute to Information on Weather Extremes and Increased Resilience

Prof. Dr. Peter Hoeppe, Former Head Geo Risks Research/Corporate Climate Centre, Munich Re Chair of Munich Climate Insurance Initiative

World Meteorological Day 2018, WMO, Geneva, March 23, 2018

How is the Business Model of Insurers Affected by Weather?

• Covering risks of loss and damage caused by weather extremes is a relevant part of the business model of insurance and reinsurance companies • In order to be able to calculate risk adequate premiums for weather related hazards they need data on the local hazards of the different weather related perils • If hazards are changing, as e.q. due to climate change, they need to quantify such trends and build them into their risk models • In order to be able to assess weather risks properly, reinsurers have started to recruit natural scientists already back in the 1970s • Reinsurers already for many decades consider global warming as a risk of change • Today also global primary insurance companies have departments with natural scientists analysing weather hazards • Insurers have created their own databases on weather related loss events Insurers as Data Providers Raising Awareness of Changing Risks

Munich Re NatCatSERVICE The world‘s most comprehensive database on natural catastrophes

The Database Today

. From 1980 until today all loss events; for USA and selected countries in Europe all loss events since 1970.

. Retrospectively, all great disasters since 1950.

. In addition, all major historical events starting from 79 AD – eruption of Mt. Vesuvius (3,000 historical data sets).

. Currently more than 41,000 data sets

Title of presentation and name of speaker 24/03/2018 4

4 2017 has been the record year for weather related losses (US$ 330 bn) Loss events worldwide - Geographical overview

Drought Winter damage, Geophysical events Wildfire Jan – Oct Flood (Western-, Southern Europe) frost (Earthquake, tsunami, (LNU Complex Fires) Jun - Oct 15 Apr - 9 May volcanic activity) 8-20 Oct South Asia Europe USA Fatalities: 1,787 Fatalities: 25 Meteorological events Wildfire (Tropical storm, extratropical storm, (Thomas Fire) convective storm, ongoing USA Flood, landslide local storm) Fatalities: 2 22 Jun - 5 Jul Hurricane Maria Hurricane Harvey China 19-22 Sep 25 Aug – 1 Sep Fatalities: 56 Hydrological events Caribbean USA (Flood, mass movement) Fatalities: 108 Fatalities: 88 Typhoon Hato 23 Aug Hurricane Irma Earthquake China, Vietnam 6-14 Sep Climatological events 12 Nov Fatalities: 22 Caribbean, North America (Extreme temperature, Iran, Iraq Fatalities: 128 drought, wildfire) Fatalities: 630 Earthquake Landslide 19 Sep 14 Aug Mexico Sierra Leone Typhoon Tembin Loss events Fatalities: 369 Fatalities: 500 22-24 Dec Philippines Cyclone Debbie Flood, landslide Wildfire (Knysna Fire) Fatalities: 164 27 Mar – 6 Apr Selection of Jan – Mar 7-13 Jun Australia catastrophes Peru South Africa Fatalities: 12 Fatalities: 147 Fatalities: 9 Source: Munich Re, NatCatSERVICE, 2018

NatCatSERVICE Loss events worldwide 1980 – 2017 Percentage distribution

Number of relevant events: 17,300 Overall losses: US$ 4,600bn Geophysical events (Earthquake, tsunami, volcanic activity) Meteorological events (Tropical storm, extratropical storm, convective storm, local storm) Hydrological events (Flood, mass movement) Climatological events (Extreme temperature, Fatalities: 1,720,000 Insured losses: US$ 1,300bn drought, forest fire)

Accounted events have caused at least one fatality and/or produced normalized losses ≥ US$ 100k, 300k, 1m, or 3m (depending on the assigned World Bank income group of the affected country).

Inflation adjusted via country-specific consumer price index and consideration of exchange rate fluctuations between local currency and US$. © 2017 Münchener Rückversicherungs-Gesellschaft, Geo Risks Research, NatCatSERVICE – As at April 2017 NatCatSERVICE Loss events worldwide 1980 – 2017 Number of relevant events by peril

Number Geophysical events (Earthquake, tsunami, volcanic activity)

Meteorological events (Tropical storm, extratropical storm, convective storm, local storm)

Hydrological events (Flood, mass movement)

Climatological events (Extreme temperature, drought, forest fire)

Accounted events have caused at least one fatality and/or produced normalized losses ≥ US$ 100k, 300k, 1m, or 3m (depending on the assigned World Bank income group of the affected country).

NatCatSERVICE Weather-related loss events worldwide 1980 – 2017 Overall and insured losses

US$ bn

Overall losses (in 2017 values) Insured losses (in 2017 values)

Inflation adjusted via country-specific consumer price index and consideration of exchange rate fluctuations between local currency and US$.

© 2017 Münchener Rückversicherungs-Gesellschaft, Geo Risks Research, NatCatSERVICE – As at April 2017 Insurance loss data provide complementing information on small scale meteorological events like convective storms

NatCatSERVICE Convective storms in Europe 1980 – 2017 Number of loss relevant events

Number

Severe storm Flash floods Hailstorm Tornado Lightning

Accounted events have caused at least one fatality and/or produced normalized losses ≥ US$ 100k, 300k, 1m, or 3m (depending on the assigned World Bank income group of the affected country).

© 2017 Münchener Rückversicherungs-Gesellschaft, Geo Risks Research, NatCatSERVICE – As at April 2017 NatCatSERVICE Convective storms in Europe 1980 – 2017 Overall losses: nominal, inflation adjusted, and normalized

EUR bn

Nominal overall losses

Inflation adjusted overall losses (in 2017 values)

Normalized overall losses (in 2017 values)

Inflation adjusted via country-specific consumer price index. Normalization via local GDP developments measured in US$ converted to EUR.

Thunderstorm related loss events end of May 2016 in Germany

Region Total losses Insured losses Fatalities Germany: esp. Baden-Württemberg € 2.5 bn € 1.2 bn 11 (Braunsbach), Bavaria

12 Thunderstorms on June 22/23, 2016 in The Netherlands

(20 mm precipitation within 10 minutes in De Bilt, humidity record for NL TD=25°C )

Hail stone in Luyksgestel (Nord-Brabant). Source: KNMI

Region Total losses Insured losses Fatalities Netherlands: Zeeland, South-Holland, € 1.3 bn € 650 mn 0 Utrecht, North-Brabant Source: Munich Re 13

Specific humidity has risen in large parts of northern hemisphere

Change in near-surface specific humidity over time in the northern hemisphere 1973–2012 (Source: Willett et. al. (2013), Clim. Past, 9, 657–677. Black dots: trends significant at the 95% level

Climate model based studies: Increase has to be expected from anthropogenic climate change (Willet et al., 2010, Environ. Res. Letter, 5; Santer et al., 2007, PNAS, 104) 14 Sea-surface temperature in tropical ocean basins with TC activity over the period1968-2016 (five-year running means)

Source: Munich Re, April 2017. Data source: Kaplan SST, via IRI, Columbia University, NYC

Sharing of data generated by insurers with society and research community NatCatSERVICE New online Nat Cat Loss Analysis Tool of Munich Re - free, no registration necessary!

http://natcatservice.munichre.com

NatCatSERVICE User Interface and Navigation 3. Chose analysis products: Number statistics, loss statistics, distributions, tables, maps

2. Select input data: Perils, regions, or filtered by (socio-)economic criteria

1. Switch between observation period or single years

4. Download results in PDF format or share it through social media

Insurers are sponsors of research projects on exterme weather

Mobiliar Lab for climate risks and natural hazards

The Swiss Mobiliar cooperative insurance company has decided to substantially expand its existing collaboration with the Oeschger Centre for Climate Research (OCCR) of the University of Bern.

A "Mobiliar lab for climate risks and natural hazards" is to be established as part of the Mobiliar Chair for Climate Impact Research in the Alpine Region at the OCCR… Other examples of research cooperations

• AXA Research Fund sponsors the Chair of Prof. Joaquín Pinto at KIT/Karlsruhe for research on extratropical storms • Cooperation between Tokio Millennum Re and Prof. Michael Kunz (KIT) in hail risk modelling project • Sponsoring of Philip Klotzbach´s (Colorado State University) hurricane forecast project by Ironshore Insurance und the Insurance Information Institute • Sponsoring of the Barcelona Supercomputing Center for analyses of hurricane forecasts by XL CATLIN insurance company • Cooperation of Munich Re with German Weather Service (DWD) and Met. Inst of FU-Berlin in project to create an event set on extratropical storms on basis of the Ensemble Prediction System (EPS) of ECMWF.

The GDV* Study: „The Climate Change Challenge“ *German Association of Insurers

Unique cooperation of science and insurance industry: • Potsdam-Institut for Climate Impact Reserach (PIK) • FU-Berlin • University Cologne

Application of a wide spectrum of climate models Dynamic as well as statistical models

Modelling of concrete loss scenarios for Germany until 2100 Model runs based on current portfolio of property insurance in Germany, results for 30 year intervals GDV*-study on future increases of nat cat losses in Germany based on climate modelling

Statistical loss model storm/hail of PIK: Regional distribution of changes in losses in a A1B-scenario relative to the average of the past 25 years

April to September (Summer)

Average change of losses in %

* German Association of Insurers 23

Insurance based researchers publish in peer reviewed scientific papers

Published in Journal „Weather, Climate and Society“ of the American Meteorological Society

The role of the insurance sector Providing recovery financing and thus increasing resilience

Insurance cover significantly helps economic recovery following a natural catastrophe: . Academic studies show that a higher level of insurance cover is accompanied by significantly better economic performance following a catastrophe. . Depending on the type of catastrophe and the level of economic development, insurance cover can even offset the negative indirect effects of natural catastrophes on national economies

. Martin Melecky and Claudio Raddatz, World Bank (2011): Higher insurance penetration at an equivalent level of prosperity leads to lower GDP losses and less government debt after natural catastrophes

. Goetz von Peter, Sebastian von Dahlen and Sweta Saxena (2012): The higher the share of insured losses to total losses, the more positive GDP performance is following a catastrophe

. Florian Englmaier, Till Stowasser (2013): The effect of insurance markets on countries' resilience: particularly in emerging economies, more insurance cover (i.e. increasing the insurance penetration rate) can mitigate the negative economic effects of natural catastrophes

Innovative parametric insurance products drive investments into new weather stations, increase availability of weather data!

G7 Climate Risk Insurance Initiative InsuResilience (2015 - 2020)

. G7 decided on a five year project to support people in developing countries to protect themselves against economic consequences of more intense and frequent extreme weather events

. Target: extra 400 million people earning less than US$ 2 per day get access to direct (100 m) or indirect (300 m) insurance of losses caused by weather extremes

. G7 Governments already pledged US$ 670 million with option of more to follow later

. Most of the new insurance products are parametric, i.e. based on the definition of triggers of meteorological parameter Parametric Nat Cat Insurance Products need Weather Data

Implementation of weather stations in developing countries more and more triggered by parametric insurance

Insurability of global warming effects in developing countries

Munich Climate Insurance Initiative (MCII) MCII Objectives of MCII: Development of risk transfer solutions to support adaptation mechanisms to global warming in developing countries within the framework of the UNFCCC process.

MCII was founded in 2005 on initiative by Munich Re together with Germanwatch, International Institute for Applied Systems Analysis (IIASA), Munich Re Foundation, Potsdam Institute for Climate Impact Research (PIK), Tyndall Centre, World Bank and independent experts. UN Climate Change Conference COP21 Paris (2015) Most relevant decisions

. Emission Reduction – limiting climate change Goal of holding global warming well below 2°C, aiming for 1.5°C

. Climate Finance Mobilizing $100bn p.a. by 2020, considerable debate over what counts as (additional) climate finance

. Climate Insurance − The Warsaw International Mechanism for Loss and Damage (WIM) introduced at COP19 to further investigate and organize the topic − Climate-related losses and damages are acknowledged as a third climate strategy pillar next to adaptation and mitigation. A clearinghouse for risk transfer will be established serving as a repository − Insurance is considered as an essential tool to address loss and damage, referenced directly under §49 of the Decisions as well as Article 8 of the Agreement

UNISDR Private Sector Alliance for Disaster Resilient Societies (ARISE) – Workstream Insurance

Commitments of partners in the workstream • Provision of natural catastrophe data • Provision of vulnerability data • Sharing expert knowledge on loss prevention • A certain percentage of investments go to risk reducing activities

Connection to 4 of 7 Sendai Framework Targets: • Reduce direct disaster economic losses • Increase number of countries with national and local disaster risk reduction strategies • Enhance cooperation with developing countries • Substantially increase availability of, and access to, multi-hazard early warning systems

Insurers Support Research for Less Vulnerable Buildings

Better Building Standards Reduce Damages

Insurance industry is the main sponsor of IBHS research institute Better Building Standards Reduce Damages

Tests to compare high-wind (160 km/h) performance of structures using common construction practices with using stronger, safer wind-resistant elements.

The components used to make the resilient building stronger and safer cost less than 5% of the total cost of the entire structure.

Wind tunnel simulation: https://vimeo.com/17764719

FORTIFIED HomeTM Tablet App This free app is a joint project between Munich Re and IBHS

• Shows step by step how to stengthen your home to better withstand severe weather events • Walks homeowners, contractors and architects through the home strengthening process, providing information based on their specific input • Information includes videos, animations, and technical specifications for building and retrofitting single family homes. Weather Risks are Top Economic Risks! World Economic Forum, Davos - Global Risk Landscape 2018

Extreme weather events Natural disasters Failure of climate-change mitigation and adaptation

Source: World Economic Forum (2018) 37

Summary

Insurance industry is:

• affected by extreme weather events in its core business • raising the awareness to changing weather risks by communicating their loss data • a provider of complementing data on weather extremes • contributing to close the gaps in the met station network • a contributor to extreme weather and resilience research