Natural Hazard Insurance in Europe - Tailored Responses To Climate Change Are Needed - Schwarze (2010)

Environmental Policy and Governance
Env. Pol. Gov. 21, 14–30 (2011)

Published online 19 October 2010 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/eet.554
Natural Hazard Insurance in Europe: Tailored

Responses to Climate Change are Needed
Reimund Schwarze,1,2 Manijeh Schwindt,2,3 Hannelore Weck-Hannemann,2
Paul Raschky,4 Ferdinand Zahn5 and Gert G. Wagner6
1
Climate Service Center Hamburg and Helmholtz Centre for Environmental Research, Leipzig
2
Leopold-Franzens-University Innsbruck
3
Alps Center for Natural Hazard and Risk Management, Innsbruck
4
Monash University, Melbourne
5
Ludwig-Maximilians-University, Munich
6
German Institute for Economic Research Berlin and Max Planck Institute for Human
Development, Berlin
ABSTRACT
This paper provides an overview of the existing systems of natural hazard insurance in
Europe, their structural characteristics and peculiarities. It also discusses the difficulties of
adaptation of these systems to climate change and the growing number of natural disas-
ters. Using Germany, Austria and Switzerland as examples, the paper demonstrates that
based on the status quo, the Swiss monopoly insurance system, which is embedded within
a unique direct voting environment, is most able to reduce micro- and macroeconomic
disruptions resulting from natural hazards. ‘Change in diversity’ is seen to offer the best
chance to arrive at insurance systems that are prepared for climate change while being
adapted to local particularities. Efforts to harmonize national and regional systems as well
as top-down EU initiatives are argued to be inadequate. Copyright © 2010 John Wiley &
Sons, Ltd and ERP Environment.
Received 7 November 2009; revised 30 July 2010; accepted 20 August 2010

Keywords: climate change; insurance; natural hazards; Europe
Introduction
W
ITHIN THE EUROPEAN UNION (EU) AND BEYOND, NATURAL HAZARD INSURANCE SCHEMES HAVE SIGNIFICANT
differences in product properties and prices. Some countries (Spain, France, Switzerland) have state or
quasi-state monopoly insurance whereas others (Germany, Italy, Poland, the UK) exert commercially
structured ‘free market solutions’, which are systematically coupled to state-funded ad-hoc relief. Other
countries (Austria, Denmark) have introduced public disaster funds financed by taxpayers’ money and still others
Correspondence to: Reimund Schwarze, Climate Service Center Hamburg and Helmholtz Centre for Environmental Research Leipzig, Leopold-
Franzens-University Innsbruck. E-mail: reimund.schwarze@ufz.de
Copyright © 2010 John Wiley & Sons, Ltd and ERP Environment
Natural Hazard Insurance in Europe 15
have various mixed solutions of private insurance providers supplemented by public disaster funds (Belgium,
Holland, Norway).
Moreover, natural hazard insurance schemes cover a wide spectrum of very different kinds of natural hazards.
They include the common ‘geo-atmospheric hazards’ (storms, hail, floods, snow load, earthquakes) and regionally
specific hazards such as landslides in the Alps or subsidence in the Mediterranean and in England. Sometimes,
even social and political risks such as civil war and terrorism, as in Spain, or general risks to buildings (fire and
burst pipes) as in the UK are covered. Furthermore, the risks are pooled differently in various countries – in some
countries storms are included but not in others, and the same holds for earthquakes.
Bearing this situation in mind, can and should one aim for a unified insurance system across Europe? The EU
Commission appears not to favour such an idea, although it does acknowledge the need for European reform.
According to the Commission, climate change demands ‘innovative solutions on the financial services and insur-
ance markets’, the ‘further integration of these solutions into the framework of EU financial services policy’ and
a ‘review of the risk structure of existing public and private disaster funds including the EU’s solidarity funds’
(Commission of the EU, 2007). The reform of natural hazard insurance is, seemingly, becoming a cornerstone
of the EU’s strategy for adapting to climate change.
Under pressure from forces within the Union in favour of de-regulation and increased competition, the EU has
previously changed the European indemnity insurance landscape, which includes hazard insurance. The 3rd ‘EU
Directive on Indemnity Insurance’ from 1992 led, for example in the then Federal Republic of Germany, to an
abolishment of all previously existing regional monopoly insurers for geo-atmospheric hazards. Other member
states such as France and Spain were more slow to change. They responded with only cosmetic amendments and in
fact have retained their national systems of insurance monopoly in this field, developed over a long period of time.
Although the EU’s efforts to unify and liberalize this sector have, for the most part, failed, there is sustained
pressure on all new insurance initiatives in Europe. Insurance initiatives such as the development of market-wide
zone tariff schemes, which the EU tends to interpret as the establishment of cartels or arrangements similar to
cartels, could be stopped by the EU Commission or the Commission’s Cartel Regulator.
The European Solidarity Fund is not neutral in the context of insurance. Although it should be restricted to
the ‘uninsured’ sectors of public infrastructure damage and aid programmes for the chronically under-insured
agricultural sector, it does have commercial links to the national and regional insurance systems. New private
regional offers such as municipal infrastructure insurance or multi-hazard insurance in agriculture may not be
able to compete as long as governmental relief is granted by the Solidarity Fund.
In the field of natural hazard insurance in Europe, there are therefore classic tensions between centralization
and decentralization, market and state, which – as in other fields of economic policy – are difficult to balance sat-
isfactorily. These tensions are played out against a backdrop of new types of economic risk caused by increasingly
extreme weather patterns, and the ‘inconvenient truth’ (Gore, 2006) is that we may not manage to halt climate
change. The only adequate response to this challenge is to increase natural disaster insurance cover and also to
increase the density of that insurance.1
We cannot cover all of the complex questions connected to these developments here but can lay the founda-
tions to arrive at some of the answers. This requires a careful stock-taking of the existing systems, including a
comparison of their strengths and weaknesses and an assessment of their ability to adapt to new conditions. This
paper aims to contribute to this process, predominantly with reference to case studies on Germany, Austria and
Switzerland, but also considering other European countries and the EU’s political initiatives.
The following section presents stylized models of natural hazard insurance. This is followed by an overview of
the existing systems of natural hazard insurance in Europe, their structural characteristics and peculiarities. This
is mainly a synopsis demonstrating the variety of the current risk transfer systems in Europe. The next section
provides reasons for the need for regionalized responses in adapting the existing diverse systems of natural hazard
insurance to climate change. We then compare three different institutions for risk transfer in the context of natural
disasters with respect to their ability to reduce micro- and macroeconomic shocks. The final section then looks
again at the issues of integration of national risk transfer systems into European policy.
1
The density of insurance is defined as the rate of insured objects (buildings) relative to the total number of objects at risk.
Copyright © 2010 John Wiley & Sons, Ltd and ERP Environment Env. Pol. Gov. 21, 14–30 (2011)
DOI: 10.1002/eet
16 R. Schwarze et al.
Figure 1. Stylized models of natural hazard insurance

(Model 1: Public Monopoly Insurer.)
Stylized Models for Natural Hazard Insurance
There are five basic models of insurance against damage caused by geo-atmospheric hazards which are distinct
from one another in terms of intensity of regulation (ex ante) and state involvement (ex ante/ex post) (see Fig. 1).
Model 1: Public Monopoly Insurer

Public monopoly insurance is the legally binding affiliation of individuals and legal entities to a specific public
insurance provider, i.e. a public monopoly insurer (in most cases, a regional one). The public monopoly insurer is
guided by statutory provisions in its contracts and public consultation processes. In practice, these insurers often
have rights of participation in public proceedings governed by public law such as planning for protection against
disasters, land usage planning and building regulations. As a result of the 3rd EU Directive on Indemnity Insur-
ance, such monopoly insurers are no longer permitted under European law. However, this does not affect the
cantonal property insurances in Switzerland and Liechtenstein. Being outside the EU, they have a unique status
as institutions of public service (Institutionen der Daseinsvorsorge) which, alongside indemnity insurance, also have
a public duty to protect from and prevent damage. This special status means that such institutions continue to be
permitted according to European law, in spite of the monopoly ban.
Model 2: Compulsory Insurance

Compulsory insurance legally regulates which categories of people, and to what extent, are obliged to purchase
insurance against (defined) damage caused by natural hazards. This compulsory purchase is almost always linked
to an obligation to contract for providers of this type of insurance, i.e. to the obligation of the insurer to offer such
insurance to interested buyers as long as they fulfil specific conditions. In this legal framework, insurance can be
offered on the market by a variety of companies. Accordingly, competition is possible to a limited extent within
the framework of compulsory insurance.
Model 3: Coupling of Contracts

The obligatory inclusion of geo-atmospheric damage into property insurance contracts (e.g. insurance against fire
or industrial insurance against business interruption) is, in effect, also compulsory insurance as the contractual
DOI: 10.1002/eet
partners are unable to decide which hazards to be insured against. However, consumer sovereignty is retained
because the parties are able to decide whether an insurance contract should be concluded at all.
Model 4: Free-market Natural Hazard Insurance with ad-hoc governmental Relief Programmes
Pure market solutions for risk transfer without state intervention do, in effect, not exist in the natural hazard field:
all unregulated systems are linked to some form of state assistance ex post in the case of extreme events. As we
know from theoretical reasoning as well as from real life experience it is difficult for the state and politicians in
office to commit not to help.2 The ‘free market’ is unable to offer extensive and comprehensive insurance against
natural hazards. This is, on the one hand, due to its limited capacity and lack of ability to control adverse selec-
tion and, on the other hand, due to the fact that private insurance of highly exposed risks is simply uneconomical.
State emergency and reconstruction aid in the case of extreme events are therefore unavoidable. This results in a
conflict of incentives: the individual decision-makers no longer engage in systematic financial risk management
due to their trust in state aid being granted.3
Model 5: Tax-financed Governmental Relief Funds (Catastrophe Funds)

Catastrophe funds provide state compensation for damage caused by natural disasters as long as the person suffer-
ing damage is not privately insured. Distinct from ad-hoc governmental relief, the funds for the compensation are
accumulated ex ante, in advance of an extreme event, based on taxation. In addition, the compensation is limited
to a defined maximum. Unlike the aforementioned models of obligatory insurance (M1–M3), a catastrophe fund
is an implicit compulsory insurance, through mandatory taxation. It is important to note that this compulsory
insurance is different from the other types in that it includes no legal entitlement to risk transfer. Although every
taxable person pays a compulsory tax contribution, if someone suffers damage, the money they receive should not
be regarded as a direct quid pro quo, but rather as aid which comes as a result of the need for help of the person
suffering damage. In addition, the tax burden is not differentiated by the degree of risks one is exposed to but by
fiscal criteria such as the ‘ability to pay’. So funds imply no incentives for relocation from risk-prone to less risky
areas, and hence no incentives are effective to tackle the problem of adverse selection.
In reality, these stylized models of insurance are often not found in their pure form, as is shown by the follow-
ing overview of some selected insurance systems in Europe.
Insurance Systems Against Natural Hazards in Europe: an Overview

Switzerland
In Switzerland, there are two independent systems of natural hazard insurance. In 19 of the 26 Swiss cantons,
public cantonal property insurers (KGVs), as monopoly institutions, offer weather-related hazard damage cover-
age, whereas in the remaining seven so called GUSTAVO cantons (Geneva, Uri, Schwyz, Tession, Appenzell IR,
Wallis, Obwalden), this protection is offered by private insurers. This parallelism of systems resulted from differ-
ences in how the EU Council Directive on Indemnity Insurance (92/49/EWG) (EEC, 1992) was adopted through
referenda in the different cantons.
In all cantons, however, there is compulsory insurance for house owners: all Swiss house owners must insure
against natural hazards and alpine risks such as storm, hail, flood, avalanche, snow loads, landslides and rock fall in
addition to insurance against fire. Regarding the terms of insurance the two systems differ considerably. The sum
insured by KGVs is unlimited and amounts to about 1.5 trillion Swiss Francs (1.1 trillion Euros) (Fischer, 2004).
Furthermore, KGVs offer coverage against earthquake risks to the amount of 2 billion Swiss Francs (1.5 billion
2
This so-called ‘Samaritan’s dilemma’ was introduced by Buchanan (1975) and was applied in a rigorous economic setting to transfer policies
in the case of extreme events by Coate (1995).
3
This crowding out of self-insurance and self-protection by government aid has been labelled ‘charity hazard’ in the literature (e.g. Browne and
Hoyd, 2000; Raschky and Weck-Hannemann 2007).
DOI: 10.1002/eet
Euros) without charging any additional premium. In contrast, in the GUSTAVO cantons the sum insured is limited
to the amount of 2 billion Swiss Francs (1.5 billion Euros) (SVV, 2006) and excludes earthquake risks. Simultane-
ously there exist deductibles to a maximum 500 Swiss Francs (370 Euros) for contents and to a maximum 10 000
Swiss Francs (7500 Euros) for residential buildings (SVV, 2006).
Buildings insurers have consolidated into two pools: one of private insurers covering natural hazard damage
in the GUSTAVO cantons and the Intercantonal Re-Insurance Group in the remaining cantons with monopoly
insurance institutions. These offer different unit tariffs covering natural hazard damage per sum insured where
the premiums in the private insurance economy are around twice that in the monopoly insurer sector. This dif-
ference in premiums has led to sustained debate about the relative economic advantages of monopoly insurance
in the field of natural hazards (Ungern-Sternberg, 2001; Kirchgässner, 2007). In addition to low advertising and
other competition costs as well as the monopoly insurers’ large reserves, which have accumulated over the years,
other reasons cited for the relative advantage of KGVs are their right to participate in the processes of the Building
Codes and Land Use Planning as well as the financing of the Fire Service and Cantonal Civil Defence
Services. In other words, there are economies of scope from the pooling of prevention and risk transfer in KGVs
(Fischer, 2008).
Spain
In Spain there is a comprehensive legal compulsory insurance against damage caused by geo-atmospheric hazards
and other ‘extraordinary events’ (terrorist attacks, political unrest). It is offered by the Consorcio de Compensación
de Seguros (henceforth, Consorcio), a state monopoly insurer. The historical roots of the Consorcio stem from
the Spanish Civil War (1936–1939). War damage in the 1930s was so great that it threatened to cause the collapse
of the entire Spanish insurance market. The Spanish government felt compelled to take responsibility for all war
damage and then to divide the costs amongst the citizens of the country through compulsory contributions. To this
end, when the war was over, the Consorcio de Compensacion de Riesgos de Motin was founded. After successfully
tackling the war damage, this institution was then expanded to cover geo-atmospheric events, natural disasters
and political unrest including terror attacks. Thus emerged the current Consorcio in 1954. Until 1990, it reported
directly to the Spanish Ministry of Finance and Economy but was then outsourced as a public monopoly institution
and since then has managed its own accounts. Distinct from private and some other public insurers, for example
in Switzerland, the Consorcio has an unlimited state guarantee in place of a conventional market-funded reinsur-
ance cover. The Spanish system survived the EU liberalization efforts (3rd EU Directive on Indemnity Insurance)
virtually unchanged. Although the Consorcio is no longer officially defined as a monopoly insurer but rather as a
‘government institution for the financing of dues for disaster damage’, everything else remains as before.
The extent to which the Consorcio is essentially still an insurer becomes clear in the method of setting of dues.
Until 1987 this consisted of a fixed percentage additional premium in the buildings, contents, accident, life and
occupational incapacity insurance branches. In the 1980s the ‘income from dues’ and the disaster damage services
of the Consorcio drifted tangibly from one another to such an extent that, for a while, there was both an under-
and then again an overcapitalisation of the Consorcio. When this happened, in the 1990s, the Spanish govern-
ment decided to make a transition to assessment of duties to the insurance values (as in conventional insurance
contracts). Since then, the fiscal charges (yearly contributions) have been as follows:
• 0.092‰ of the insurance sum for buildings
• 0.18–0.25‰ of the insurance sum for interruption of operation for business or industry risks
• 0.35–2‰ of the insurance sum for public infrastructure (motorways, ports)
• 0.096‰ per person in accident insurance
• 4.45 Euros per vehicle in car insurance.
These ‘charges’ are raised by private insurers in Spain and passed on to the Consorcio. To collect these funds, they
receive remuneration based on their costs.
Officially, in Spain, private insurance companies can also offer insurance cover against geo-atmospheric damage
(which is why there is no violation of EU Competition Law) but, given that these companies also need to include an
additional premium for the Consorcio, households insured through a private company would have to pay double
DOI: 10.1002/eet
for their atmospheric damage cover. This ‘potential private competition’ is therefore insignificant. Furthermore,
private insurers do not have access to a state guarantee, which means it must be more expensive than the Consorcio.
The density of insurance is high, due to the obligatory nature of natural hazard insurance. It is dependent on the
density in the individual insurance branches –for example, in the field of non-life insurance this is around 70–80%.
Those taking out insurance must pay an excess usually of around 10% of the damage sum (at least 150 Euros)
but this excess is again dependent on the total insurance, limited to a specific amount (1%).
France
The French natural hazard insurance model is, in many ways, similar to the Spanish model but includes certain
country-specific market-economy and state-control elements. First, since 1982, all private insurers in France
(domestic and foreign) are obliged by law to provide comprehensive insurance protection against natural hazards.
This insurance protection is mandatorily tied to various property insurance contracts. Every customer purchasing
property, car or business interruption-of-operation insurance is obliged to purchase this type of cover or to forgo
property insurance altogether. At the same time, the further treatment of the risks is clearly differentiated between
market ‘insurable’ risks (storm, frost, hail, snow load) and market ‘uninsurable’ natural disasters (natural catastro-
phes, henceforth abbreviated as CatNat4). Market insurable risks remain for the most part in the normal business
of the insurers and reinsurers whereas CatNat come under a special insurance model that is legally regulated. The
government takes on the functions of the insurance economy in this case. This begins with determining whether
an event can be defined as a natural disaster. Although natural hazards in Spain are clearly laid out in laws and
contracts, in France, a governmental commission consisting of Members of the Home Office and Ministry for
the Economy and the Environment decides on a case by case basis whether the conditions for a natural disaster
are fulfilled. This has recurrently meant that damages in the CatNet model that were previously not at all or not
entirely taken into account have been covered – most recently the damage to properties caused by subsidence as a
result of the summer drought of 2003 in southern France. In the history of the CatNat model, this has repeatedly
led to a critical situation at the ‘Caisse Centrale de Reassurance’ (CCR). The CCR is a state reinsurance institution
which offers private insurers the opportunity to buy insurance against natural hazards under special subsidized
conditions. For this, the French government gives the CCR an unlimited government bond and special tax exemp-
tions for the treatment of surpluses in insurance business. Formally, primary insurers can also purchase their
insurance on the conventional reinsurance market but they would receive far worse conditions and pay more than
if they purchased from the CCR.
For the mandatory CatNat, which is legally coupled to all property insurance contracts, a uniform supplement
of 12% of the insurance premium (6% in the case of car insurance) must be collected. As a consequence, in
France, primary insurers were able to adjust the basic premiums for their property insurance products by region
and thus to achieve a selection of ‘good risks’ in their portfolio. Or they have outsourced regions or sectors which
are particularly susceptible to damage to a special insurer, which then transferred all risks to the state CCR. This
resulted in an agglomeration of ‘bad risks’ (adverse selection) at the CCR. Today, the retention rate, i.e. the rate
of excess of the direct insurer, is fixed by law. The excess sum to be paid by insured persons is also fixed by law.
Excesses are comparably low, for example 380 Euros each for damage to property and cars, but are unalterable to
maintain the insurer’s incentive to prevent damage. The excess sums multiply for recurrent damage and a lack
of preventive measures by the municipality. The aim of this stipulation is to increase the self-protection efforts of
insured parties and public institutions.
Belgium
Belgium is rarely affected by natural hazards. The most frequently occurring natural hazards are storms, floods
and, to a very limited extent, earthquakes. Significant losses were most recently experienced in January 1990
during storm ‘Daria’, which caused damage estimated at over US$1 billion. Significant damages were also caused
by floods in 2002.
4
http://www.CatNat.net
DOI: 10.1002/eet
Since the 1970s, Belgium has had a disaster fund which provides staggered state compensation of up to 250,000
Euros with a very low excess (250 Euros) for all damage caused by extreme events. This fund has, however, proven
cumbersome in practice with long waiting times for those affected and payments are impossible to calculate because
the ‘insurance case’ was not defined but rather, as is still the case today in France, depended on the declaration of
a disaster by the Belgian government.5 Due to an extreme event in 1990, the need for a new system which allows
calculable risk transfers and gives more obligations to private insurers increased. The result was the establishment
of an obligatorium, which included both insurance against natural hazards and social conflicts (terrorism, political
unrest). Initially, this obligatorium included only ‘classic’ natural hazards (storms, hail, snow load). After the extreme
floods of 2002, by royal decree the consequences of flooding, high ground water, earthquakes and subsidence were
also included. The disaster fund remains subsidiary and intervenes when no private coverage is available or private
coverage would be too expensive, for example for the agricultural sector. This is therefore a so-called ‘Public–Private
Partnership’. The premiums for the obligatorium are low (1.5‰) and are subject to a tariff for any supervisory board’s
oversight. The government offers a guarantee for private insurance of up to 280 million Euros per insurer and event
if the damage per insurer and event exceeds a maximum of 3 million plus 0.35 times the premium income.6
The Netherlands
The Netherlands experiences frequent natural hazards, mainly storms, extreme rainfall and droughts. Until the
1990s, however, it was not possible to purchase private insurance against floods and earthquakes. Insurers, toler-
ated by the Dutch government, agreed not to offer flood or earthquake insurance. For floods caused by sea water
(storm floods) and floods due to the collapse of river dikes, this agreement remains valid. Other sectors were
transformed under pressure from the European competition regulation (Commission Regulation EWG 3932/92 of
21/12/1992) in a ‘non-binding recommendation’ by the insurance sector. Since 2004, insurance for damage due
to extreme rainfall is generally offered. In practice, however, no national insurance market for flood hazards has
emerged. Under increasing public pressure after the Roermond earthquake (1992) and the flooding of the River
Maas and Rhine (1993 and 1995) the Dutch government was forced to establish legislation [Wet Tegemoetkoming
Schade bij Rampen en Zware Ongevallen (WTS)]. This legislation can be used by the central government in rare
cases, to (partly) compensate losses from events that were not insurable or covered in any other way, including
both natural and technological accidents. Compensation does not come from a fund, but from ex post payments
from the current budget of the national government.
Government compensation from the WTS has been provided during a number of extreme events, such as the
heavy rains of 1998 and 2002 and the dike failure in Wilnis in 2003. At the time of writing, insurance against
rainfall damage is not yet available. The WTS legislation leaves room for the definition of a ‘disaster’, and, in a
number of recent damaging events, the legislation was deemed to be not applicable by the Minister of Internal
Affairs. At the same time, the government attempts to limit its liability, causing some movement on the Dutch
insurance market. For example, a Dutch Agricultural Insurance Pool (Agriver, LTO Aquapool) was established,
which provides coverage for loss of crops due to heavy rains if the WTS guarantees compensation for large catas-
trophes. Moreover, since 2004, damage due to extreme rainfall is often covered in content policies for households.
For all other sectors, however, the market is stagnant, which means that regular ad-hoc governmental relief through
the TWS legislation continues to be provided.
Germany
In Germany, there is a pure market-based insurance system with individual premium calculation based on risk
exposure. Insurance against storms and hail is prevalent. However, the insurance density against other natural
5
The process is, however, still more based on regulation than in France. In order for an event to be classed as a disaster, it must be of unusual
occurance and lead to considerable damage. The following criteria apply: total damage must amount to at least 1239,467.60 Euros, average
damage per family must total 5577.60 Euros and, furthermore, the event can only occur once in 20 years (according to a ministerial circular
of 30 November 2001).
6
For earthquakes, the sums are higher: 700 million Euros per insurer and an event with damage exceeding 8 million Euros plus 0.84 times
the premium income.
DOI: 10.1002/eet
hazards is very low (less than 10% for buildings and contents). If extreme events occur, ad-hoc relief is often
provided for emergency and reconstruction, the latter predominantly in the form of tax exemptions. Victims of
damages do not, however, have a legal right to these programmes, which are subsidiary to the provisions of private
insurance. This means it will only be provided to those who could not take out private insurance.
For events generating extensive media coverage, public relief generally takes over. For example, in the Elbe and
Danube floodings of 2002, (then) chancellor Gerhard Schroeder, who was running for re-election, promised gov-
ernmental emergency assistance amounting to 400 million Euros and during visits to the flooded areas assured
victims full compensation for damage, i.e. that ‘nobody will be worse off after the event’. Claimants in this case
could expect and in fact did receive a high level of compensation resulting from this ad-hoc scheme (Schwarze
and Wagner, 2004, 2007).
Poland
Poland is mainly threatened by floods. Large river valleys such as the Wisla and the Odra, which are well devel-
oped and densely populated, are frequently affected so Polish exposure to flood risk can be considered to be high
(Ostrowski et al., 2002).7 Recently, Poland has been hit by disastrous floods in 1997, 2002 and 2010. Direct property
losses from the 1997 flood have been estimated at about 2.7% of Poland’s GDP (Kunreuther and Linnerooth-Bayer,
2003).8 Insurance coverage against floods and other natural hazards is available in Poland on a voluntary basis for
residents and enterprises. The take up of storm coverage is widespread (CEA, 2005) and the density of flood insur-
ance is about 25% for private households and between 25 and 50% for business and industry (Bouwer et al., 2007).
Nevertheless, the government intervened in both major floods in 1997 and 2010 with ad-hoc relief for flood victims.
Those who suffered direct losses by the flood of 1997 were offered financial aid of 3000 zlotys (equivalent to 700
Euros) or three times the average net monthly salary (Kundzewicz et al., 1999). In the most recent flood event of
2010 the Polish Prime Minister Donald Tusk promised immediate financial aid – a lump sum of 6000 zlotys (1500
Euros) – to flood victims (thenews.pl, 2010). Against this background, Poland has – similar to Germany – a public
ex-post compensation system for disaster losses which is ad-hoc funded by tax revenues (see also Kunreuther and
Linnerooth-Bayer, 2003; Aakre et al., 2010). But there is an important difference to the compensation scheme in
Germany: in Germany only uninsured victims are compensated by public reliefs. In Poland, every flood victim
receives relief irrespective of holding insurance or not (Bouwer et al. 2007). As every insured and uninsured victim
received lump sum relief to cover a part of his or her losses, no crowding out of private demand for insurance should
be expected.9 Another difference between Poland and Germany is that whereas private and public flood losses have
often been overcompensated in Germany (Citlak and Wagner 2001), a large share of direct losses in Poland remained
uncovered in 1997 and later events.10 Overall the current Polish risk management is characterized by a high degree
of uncertainty. Victims can neither rely on governmental relief (as funds may turn out to be insufficient) nor expect
to receive sufficient compensation to repair damage (as insurance coverage is often restricted).
Austria
In Austria the private supply of flood insurance is supplemented by a tax-based catastrophe fund. The Austrian
Catastrophe Fund Act was established in 1966 after severe flood events occurred in 1965 and 1966 in order ‘to
7
Bouwer et al. (2007) consider Poland, the Czech Republic and the Netherlands to be particularly vulnerable to flood risk as losses as a result
may reach more than 2% of GDP.
8
The flood of May 2010 has been an extreme event. According to the Polish Prime Minister Donald Tusk the flood has been ‘the worst natural
disaster in the nation’s history’ (RTE News, 2010). At least 25 people died, approximately 23,000 people were evacuated and economic losses
have been calculated at least up to 2.5 billion Euros (thenews.pl, 2010). However, the flood of 1997 with 54 fatalities and 162,000 evacuated
people, was even more dramatic. Economic losses of the 1997 flood event are estimated at 2–3 billion Euros (Kundzewicz et al., 1999).
9
If only uninsured victims are able to claim partial relief, there is an incentive to set private insurance aside in order to save the premiums for
the insurance and rather rely on costless relief. As every insured and uninsured victim in Poland received partial compensation to their losses
from the government there is no need to substitute private insurance coverage for governmental relief. If an insured victim receives the same
amount of relief as an uninsured person, the incentive for buying private coverage should not change.
10
The Polish government spending covered about 43–48% of total direct losses in 1997 (Kunreuther and Linnerooth-Bayer, 2003; Aakre et al.,
2010). And 8–10% of direct losses were covered by insurances (Kunreuther and Linnerooth-Bayer, 2003; Aakre et al., 2010). Thus, the large
remaining share of direct losses (about 44–46%) was not covered.
DOI: 10.1002/eet
provide financial support for victims of natural hazards’. To ensure the financing of the fund, a mark-up on income
tax, capital yields tax and corporate income tax was levied. This generates, depending on the economic situation,
fluctuating takings. Until 1996 large reserves were built due to the absence of severe natural hazard events. In
order to increase the fungibility of these reserves special statues were implemented. For example, the new Catas-
trophe Fund Act limits the reserves to a maximum of 29 million Euros. In cases of exceeding costs due to extreme
events, additional means were provided by the government. For the flood events in 2002 and 2005, for example,
the government allocated an additional 500 and 251 million Euros, respectively. As the tax-based catastrophe fund
is in fact a form of ‘enforced solidarity’, its financing can be interpreted as a quasi-compulsory insurance. However,
as opposed to insurance contracts, the taxpayer does not have any legal rights to a risk transfer in the case of
damage. For this reason the ex-post transfer should not be considered as indemnification payment but rather as
relief. Once a flood event occurs, the person concerned has to apply for governmental relief. Because the ultimate
authority for all decisions concerning the catastrophe fund rests on the respective Federal States with each State
using its own directives, varying levels of governmental relief can be observed for comparable cases across Federal
States (Prettenthaler and Vetters, 2005, Albrecher and Prettenthaler, 2009).
Overview
These and other national systems of natural hazard insurance in Europe are summarized in Table 1. The existing
European systems of risk transfer are sorted according to the stylized type systems described above cum grano
salis. The result is shown in column 3, with mixed and transitional types being indicated by a slash (Mi/Mj, i ≠ j).
The overview given in Table 1 clearly demonstrates the heterogeneity and diversity of the systems of risk transfer
for natural hazards in Europe. Against this backdrop, it seems obvious to question whether it makes sense at all
to aim for standardization or collaborative European efforts in the field of natural hazard insurance in the context
of climate change. Our arguments below suggest that the answer should be no.
Regional Responses to Climate Change are Needed
The natural hazard situation in EU member states varies from a climatic, topographic and institutional perspective.
In Austria and Switzerland, with the highest levels of regulation or government involvement, almost ten times the
number of people live in risk zones than in Germany (Url and Sinabell, 2008). Hazards in the high alpine valleys
during the ‘same’ natural events (heavy rains) are completely different from those in lowland areas even within one
geographical region such as Tyrol or Vorarlberg (Pfurtscheller and Schwarze, 2008). There is practically no early
warning time and few retention areas to give the ‘rivers space’. There are also differences, developed over the years,
in building methods, for example many more wooden houses in the Alps and practically no brickearth buildings
as in the north European lowlands. The list of special regional and even local vulnerabilities goes on. Off-the-peg
coverage here does not make sense. It may even destroy the human and social capital which has developed over
time to combat natural hazards in Europe’s regions. For example, the ‘solidarity’ system of cantonal monopoly
insurance in Switzerland is deeply rooted in the social perception of belonging to one ‘risk community’ where
members share a collective exposure (Fischer, 2004).
Despite remaining uncertainties, it clear that, as a result of global climate change, natural hazards will increase
in severity and occur more frequently almost everywhere (IPCC, 2007, table SPM.2 for a worldwide overview;
Beniston et al., 2007 for Europe). Today, events which occur once a century will be something we might see
once every 20 years in the future (Allamano et al., 2009; Kron, 2010, p. 154). From region to region, however,
this effect varies significantly.11 Furthermore, climate change has a differential impact from region to region. On
the one hand, in southern Europe, we can expect declining flood hazards but, at the same time, more frequent
occurrence of droughts; in northern Europe, on the other hand, we can expect more flooding. In alpine areas the
11
Allamano et al. (2009, L24404) study return period ratios (RPRs) for different scenarios of climate change (in 2100) and different precipitation
scenarios in Switzerland. The RPRs range from 1 to 19.1, implying ‘no change’ of current return periods (under constant regional temperatures)
to a 20-fold increase (under a 5°C increase in regional temperature).
DOI: 10.1002/eet
Country Description Model
Switzerland Dual system of private and public insurance with monopoly character. In all cantons, fire M1/M2
insurance and insurance against atmospheric damage is mandatory for all buildings and
household effects (value as new) with an excess of 10% per incident of damage or at least
200 CHF with a maximum value of 1000 CHF (approx. 680 Euros). Reinsurance is provided
via two pools of direct insurers with compulsory membership. The pool system for cantonal
property insurance offers unlimited cover whereas the private insurance pool for atmospheric
damage only provides coverage for up to 25 billion CHF (17 billion Euros). The public
insurers and the KGV link the risk transfer with the maintenance of the emergency services
(fire service) and have the right to participate in Federal State Planning and Land Use
Planning. Insurance density is almost 100% in Switzerland.
Spain There is a legal obligation to insure against damage caused by natural hazards and other M2/M3
‘unusual events’ (terrorist attack, political unrest). Premiums are collected by private
insurers as an additional premium in the following sectors: building, contents, accident,
life and occupational incapacity insurance and are passed on to the so-called Consorcio de
Compensación de Seguros (Consorcio), which is a state monopoly insurer. The Consorcio is
subsidized by an unlimited government guarantee. Private insurance companies can formally
offer competitive natural hazard insurance coverage. However, as these companies also
have to pass on the additional premiums to the Consorcio, this competition is, in practice,
insignificant. The insurance density is high, depending on the density in the individual
sectors, up to 80%. Insurers’ excess is usually around 10% but is limited depending on the
insurance sum.
France Mandatory inclusion of all ‘uninsurable’ natural hazards (not including storm, frost, hail M2/M3
and snow load) in all contents insurance contracts by way of a uniform surcharge of 12%
on the insurance premium with a low excess (e.g. 380 Euros per incident of damage to
buildings and cars). Reinsurance is offered at a fixed cost through the state Caisse Centrale
de Réassurance (CCR) with an unlimited state guarantee. Private insurance companies can
formally reinsure but, as this is not subsidized by the state, the insurance they offer is more
expensive than reinsurance through the CCR, which thus in reality has a monopoly over
reinsurance. High insurance density of close to 100%.
Belgium Mandatory inclusion of all natural hazards in compulsory insurance linked to fire insurance. M3/M5
Private coverage occurs through a direct insurer. In the case of some defined extreme
events, a state disaster fund subsidizes. The disaster fund also covers damages that are
not insurable on the private market, particularly in agriculture. The tariffs are regulated
(insurance supervisory board) and are low (1.5% of the total insurance sum) with minimal
excesses (250 Euros).
Great Britain Pure private insurance with risk-based individual premium calculation, i.e. high supplementary M3/M4
premiums for high exposed risks. High market penetration (75% of private buildings,
95–100% of mortgage credits) through integration of natural hazard insurance into fire
insurance, which is widely required when taking out a mortgage to secure credit.
Germany Pure private insurance with individual premium calculation in the case of flood damage based M4
on risk zonings. Insurance against storm and hail is prevalent (95%). However, insurance
density against other natural hazards is under 10%. German banks regularly require fire
insurance for mortgages but no insurance against natural hazards. If an event occurs –
particularly in the case of large events which are extensively covered by the media such as in
2002 – ad-hoc relief is often provided for emergency and reconstruction. Victims of damage
do not, however, have a legal right to this governmental relief, which is subsidiary to the
provisions of private insurance.
Italy Insurance against natural disasters is fully private without any government regulation. There are M4
ad-hoc governmental disaster funds financed by taxpayers ex post. Victims of natural hazards
do not have a legal right to access these funds. Other state reliefs are assigned after specific
events at the government’s discretion.
Continued
DOI: 10.1002/eet
Country Description Model
Poland Private insurance market on a voluntary basis. Mandatory insurance only in agriculture. M4
Insurance density for storms and floods is relatively high. Flood insurance density is
estimated to be about 25–50%. Hence, there still exists a considerable number of uninsured
victims. The Polish government provides ad-hoc relief with lump sum transfers. A major part
of the losses remain uncovered as there has been a significant lack of private and public
financial resources. Overall the current Polish risk management is characterized by a high
degree of uncertainty.
Netherlands There is effectively no insurance protection against damages caused by large-scale flooding, M4/M5
based on a ‘non-binding recommendation’ by the Dutch insurance sector with one exception:
loss of harvest due to heavy rains is covered by a pool of insurers for the agricultural sector.
A government disaster fund covers all ‘uninsurable’ damages. Frequently, however, there is
additional governmental ad-hoc relief in the case of extreme events. Legislation is currently in
the process of being updated.
Austria Insurance against storms, hail and snow load is usually provided by way of private contracts M4/M5
without governmental regulation. Private additional coverage against atmospheric damage
is possible but is rarely used. Insurance density for protection against these natural hazards
(flooding, avalanche, landslides, etc.) is under 15%. Since 1986 Austria has had a government
disaster fund under the Finance Ministry financed by taxpayers. Although victims of damage
do not have a legal right to access this fund, it can cover approximately 50% of damages
(on average) if the claimant is not privately insured at the same time. This leaves room for
‘charity hazard’ in the case of private insurance.
Table 1. Natural hazard insurance in Europe

Sources: Ungern-Sternberg (2002), Michel-Kerjan (2001), Kunreuther and Linnerooth-Bayer (2003), Lasut (2003), Huber and
Amodu (2006), Prettenthaler and Vetters (2005), CEA (2007), Raschky and Weck-Hannemann (2007), Schwarze and Wagner
(2007), Bouwer et al. (2007), Url and Sinabell (2008), Bruggeman et al. (2008), Fischer (2008), Huber (2008), Raschky
et al. (2008), Albrecher and Prettenthaler (2009), L. M. Bouwer and D. Porrini (personal communication).
risk of rockfalls will increase with the melting of glaciers; in southern Europe, the risk of subsidence will increase
as a result of soil dehydration. Moreover, due to special regional vulnerabilities the ability to adapt to this trend
also varies. Harmonization or joint EU efforts are unlikely to be able to cater for the needs of such a variety of
problem zones and solution strategies. We certainly need to improve awareness of natural hazards across Europe
and careful steps need to be made in national policy to adjust systems of natural hazard insurance, considering
regional and local vulnerabilities and coping capacities.
The following section will show that European systems of risk transfer for natural hazards respond very dif-
ferent to similar risk exposures, providing different degrees of coverage at different speed. Under given local
circumstances even public monopoly insurance for natural hazards can be more efficient in this respect, if it is
locally well adapted.
Effectiveness of Risk Transfer Mechanisms in Austria, Germany and Switzerland: Results from
a Case Study12
In August 2005 parts of Austria, Germany and Switzerland experienced a major flood event with enormous finan-
cial losses.13 As already discussed, these countries apply very different risk transfer systems in order to absorb the
adverse effects of natural hazard events:
12
For detailed information on the case study, see Raschky et al. (2008).
13
Total damage amounted to approximately 1.42 billion Euros (2.2bn CHF) in Switzerland, 592 million Euros in Austria and 180 million Euros
in Germany.
DOI: 10.1002/eet
Figure 2. Sum of precipitation (l/m2)

(Source: Department of Meteorology and Geophysics, University of Vienna, 2005.)
• M1: Regional public monopoly insurance (in the non-GUSTAVO cantons of Switzerland)
• M4: Pure market solution with supplementary ad-hoc governmental disaster relief (Germany)
• M4/M5: Tax-based disaster fund with supplementary market insurance (Austria).
The flood event in August 2005 in the northern Alps was caused by a series of intensive precipitation events, which
reached their climax in an extreme rainfall event of 22 and 23 August. The trigger for these extreme precipitation
events was a so-called Vb weather pattern with a Mediterranean low-pressure area (Norbert) at its centre. In the
course of its journey across the Gulf of Genoa, Norbert was saturated with water vapour and then flowed across
the Balkan region in from the north-west towards the Alps, where it was forced to rise and cool. From 22 August
this led to relief rainfall, which was intensified by local fronts. Several precipitation cells overlapped one another
and led to rainfall across the entire region of the northern Alps. The intensity of the rainfall was highly variable in
terms of both location and time, although the cumulative maximum was located exactly in the areas of Grisons,
Vorarlberg, Tyrol and in the Alpine foothills of Bavaria (see Figure 2, red shading).
The unique constellation of three countries, each having a different institutional design to cope with the after-
math of natural hazards, being hit by the same event allows us to compare these risk transfer mechanisms with
respect to their ability to reduce micro- and macroeconomic shocks due to natural hazards under relatively similar
conditions. Henceforth this comparison will be based on three affected sub-regions: Upper Bavaria (Germany),
Tyrol (Austria) and Grisons (Switzerland).
Considerable flood events often result in interruptions of production processes, but also in declines of con-
sumption, whereby both consequences can affect the macroeconomic development negatively. In order to smooth
disruptions due to natural hazard events and to foster reconstruction, a comprehensive, quick and effective risk
transfer is required. This includes the reduction of misguided incentives: a guaranteed insurance coverage might
reduce the incentive for protective measures (moral hazard ex ante) and to limit damage to a degree not avoidable
(moral hazard ex post).14 Moreover, the promise of governmental or private assistance might crowd-out the demand
for natural hazard insurance (Raschky and Weck-Hannemann, 2007).
14
The distinction between moral hazard ex ante and ex post goes back to Zweifel and Eisen (2003, pp. 295–320).
DOI: 10.1002/eet
Tyrol Upper Bavaria Grisons
State measures 53.3* 22.0* 100†

Private measures 12.8 18.3
Table 2. Damage coverage (%)

Note: * transfer at current value; † transfer at replacement value.
Data
In order to compare the three different institutional designs with respect to their ability to absorb micro- and mac-
roeconomic shocks, surveys conducted in Tyrol and Vorarlberg15 as well as Upper Bavaria were used. The former
contained 218 households, 72 of which were hit by the flood event in August 2005. The remaining 147 households
were used as a control group. The second survey included 305 affected households from Upper Bavaria. All data
necessary for the third investigation area – Grisons – were provided by the KGV Prevention Foundation.
Damage Coverage
Before investigating which of the three institutional designs can provide the most comprehensive damage coverage,
it is important to point out that compensation in these three systems differs in two major aspects:
• Whereas damage claims in the public monopoly system of Grisons are compensated at replacement value,16
compensation in Tyrol and Upper Bavaria is limited to current value with additional provisos.
• In contrast to Grisons, where compensation of the monopoly insurer is based on a legal claim, governmental
transfers in Tyrol and Upper Bavaria cannot be interpreted as such. Rather, they have to be seen as governmental
relief provided on a voluntary basis.
Table 2 summarizes damage coverage by private insurers and the government in the areas investigated.
By May 2006 the government of Tyrol received 920 claims from private households and companies related to
the flood event of August 2005. On average, 53.3% of the damages were covered by the Austrian catastrophe fund.
According to the government of Tyrol, 12.8% of approved damages were financed by private insurers. In 515 of
920 reported cases, private insurers provided some coverage. The average indemnification payment among those
who were insured was on average 25.4%.
Similar to the situation in Tyrol, the government of Upper Bavaria provided relief for flood victims. However, the
two systems of governmental assistance differ in two aspects, both of which might impact the problem of charity
hazard: (1) in contrast to the Austrian catastrophe fund, governmental relief in Germany is not institutionalized
and is therefore unreliable (Raschky et al., 2010) and (2) governmental relief in Upper Bavaria was only provided
to those who did not suffer an insurable damage.
Private households received compensation of at least 500 Euros per person up to a maximum of 2500 Euros,
given a minimum damage of 5000 Euros which was not insurable. Private households with oil damage were
compensated with an amount up to 5000 Euros. Determination of the damage coverage due to governmental
action is based on a sample of applications for disaster assistance by commercial enterprises in the districts of
Garmisch, Bad Tölz, Munich district and Erding. As this sample is not representative, the results should be inter-
preted with caution. There is a wide of insurance coverage, resulting from the differences in insurability in each
district: whereas damage coverage by public relief programmes in an uninsurable region such as Garmisch (41%)
is very high, the value is very low for Bad Tölz and Munich district (15%) where flood damage can be insured. The
average damage coverage resulting from governmental action is 22%.
15
As many affected people in Tyrol (Austria) did not answer, probably because they did not want to be reminded of the disasterous event, it
was necessary to expand the survey to Vorarlberg (Austria).
16
Replacement value (value as new) refers to the outlay of costs necessary to erect a building of a similar kind, size and structure.
DOI: 10.1002/eet
Mean 53 38 14
90% of claims 119 135 21
Table 3. Time lag (days)
>1* >1† 0.3†
Table 4. Premiums (‰)

Note: * fixed sum insurance; † not actuarially fair.
Furthermore, the Bavarian Chamber of Insurance, which provides building insurance to 85% of all house
owners in Bavaria, paid 5.5 million Euros to its customers. Taking into account the total damage of 30 million
Euros for private households and commercial enterprises in Upper Bavaria, this corresponds to damage coverage
by insurers of 18.3%.17
The monopoly insurer in Grisons is obliged to compensate damage at replacement value, resulting in 100%
damage coverage less excess.
Time Lag
In order to reduce economic disruptions due to natural hazards, a rapid response of the authority or insurance
company is necessary. The quicker claims are dealt with and financial compensation is paid, the less insecure the
situation of affected people will be, allowing them to return to business as usual and keep the direct economic
effects of the disaster event relatively small. Table 3 shows the time lag between the date of claim and the approval
of the damage. Note that comparability of the results is limited due to the different origins of the data.18
The Austrian catastrophe fund needed on average 53 days for approval of the damage. After 119 days about 90%
of the claims were verified.
As official data for Upper Bavaria are not available, we used the questionnaire described above as a basis for our
comparison. According to this questionnaire, the time lag between the registration and approval of the claim was
on average 38 days. After the verification, another 37 days passed until the interviewees were compensated. Some
90% of the claims were approved within 135 days.
According to information of the KGV Prevention Foundation, an average of 14 days is needed for the determina-
tion of damages, and after 21 days compensation at current value is paid. The difference between current value and
replacement value is paid on the basis of the reconstruction work. However, as both the Austrian Catastrophe Fund
and the German ad-hoc system compensate at current value, these are the relevant figures for the comparison.
Result 1. Micro- and macroeconomic disruptions are better absorbed by public monopoly insurance than by
disaster funds or ‘free market’ insurance with ad-hoc governmental relief.
Premiums
Thus far, our comparison of the different institutions for risk transfer after natural hazards has focused on the ben-
efits of the systems, thereby neglecting the costs. Table 4 compares the premiums for the three types of insurance.
17
However, it is important to note that only 700 insurance holders reported a financial loss due to the flood events, i.e. the amount of 5.5
million Euros was paid to these 700 affected insurance holders.
18
Whereas data for Tyrol and Grisons are provided by the government of Tyrol and the KGV Prevention Foundation, respectively, figures for
Upper Bavaria are based on our survey.
DOI: 10.1002/eet
In addition to fire insurance, people in Tyrol can purchase natural hazard insurance. The yearly net premium of
a fixed sum insurance for a house worth 335,000 Euros would be approximately 420 Euros, i.e. approximately 1‰.
The calculations for Germany are based on the supply of two insurers. The first insurer (Bruderhilfe) provides
insurance coverage for fire and natural hazard damage (i.e. floods and storms). Assuming a house worth 300,000
Euros and an excess of 1% of the sum insured, the yearly net premium would amount to 313 Euros. In contrast, the
price of the second insurer (Gerling) is based on an excess of 10% of damage. In this case the yearly net premium
is 376 Euros. These figures translate to a relative premium of more than 1‰ for insurances in Germany.
As already discussed, the monopoly insurer in Grisons can provide insurance coverage for a house worth
500,000 CHF (∼335,000 Euros), for a yearly premium of 150 CHF (∼100 Euros). This means that the relative
premium of an insurance supplied by the KGV is 0.3‰, i.e. less than one-third of the premiums in Austria or
Germany.
Result 2. Public monopoly insurance is cheaper due to cross-subsidization.
Case Study Résumé

As this case study is based on a wide range of data sources, the results need to be interpreted with caution. Nev-
ertheless, the outcome of the comparison is clear: compared with the Austrian catastrophe fund and the German
ad-hoc system, the public monopoly in Grisons performs better in reducing micro- and macroeconomic shocks.
The indicators – damage coverage and time lag – are clearly in favour of the public monopoly. The advantages of
the public monopoly are not only to be found on the benefit side, but also on the cost side. Due to cross-subsidizing
and lower transaction costs, the monopoly insurer in Grisons can supply insurance coverage for less than one-third
of the premium of comparable insurances in Austria or Germany.
But these advantages come at a price: this system of public monopoly (1) delimits consumer sovereignty and (2)
prevents competing companies from providing the same product. In fact, this limitation of consumer sovereignty
seems to be the price for an efficient system of risk transfer. On the other hand, this system of public monopoly
in the Swiss canton of Grisons is connected with far-reaching voting rights of the citizens in a direct democracy.
For this reason the restrictions of consumer sovereignty might be compensated by voter sovereignty. And not
least, the results on the election for a partial revision of the law governing building insurance on 26 November
2000 show that the people of Grisons are satisfied with the current form of risk transfer (86% yes versus 14% no
votes).
Conclusion
The overarching question being addressed in this paper is: should we aim at a unified insurance system for natural
hazards in Europe? Our case study shows that European countries differ in many respects. We see that – under
given local circumstances – even a public monopoly insurance for natural hazards can be more efficient. Insurance
systems in Europe have developed over long periods of time and are partially national, partially regional institu-
tions which, to a great extent, are adapted to the natural and socio-historical conditions of the regions they cover.
The insurances are rooted in the different cultures of the societies in combating natural hazards, and sometimes,
as for example in Switzerland, they are deeply rooted in the collective ego and a specific framework of political
institutions. Calls for a standardization and optimization have to be evaluated according to the extent to which
they take these differences in the natural and social conditions into consideration.
But this is not a plea for maintaining the status quo. If changes in natural conditions resulting from climate
change or changes in social conditions – such as a further growth in population and accumulation of assets in risk
zones – make it necessary to move towards reform then these adjustments are unavoidable. In some countries,
such as Switzerland, the changes are already in motion. Other countries such as Germany, Italy and the Nether-
lands are lagging far behind. The obstacles facing system change are numerous across the board (Schwarze and
Wagner, 2007). The adjustments to new weather conditions are likely to reflect the existing differences in the
regional and national insurance systems in the EU in some way.
DOI: 10.1002/eet
In this context, ‘change in diversity’ offers the best chance across Europe to achieve systems that are effective
and at the same time adapted to local particularities. Efforts to harmonize national insurance systems and EU
initiatives should be rejected in this phase of uncertainty and necessary learning.
References
Aakre S, Banaszak I, Mechler R, Rübbelke D, Wreford A, Kalirai H. 2010. Financial adaptation to disaster risk in the European Union –
Identifying roles for the public sector. Mitigation and Adaptation Strategies for Global Change doi: 10.1007/s11027-010-9232-3
Allamano P, Claps P, Laio F. 2009. Global warming increases flood risk in mountainous areas. Geophysical Research Letters 36: L24404.
doi:10.1029/2009GL041395.
Albrecher HJ, Prettenthaler F. 2009. Hochwasser und dessen Versicherung in Österreich. Austrian Academy of Sciences Press: Vienna.
Beniston M, Stephenson D, Christensen O, Ferro C, Frei C, Goyette S, Halsnaes K, Holt T, Jylhä K, Koffi B, Palutikof J, Schöll R,
Semmler T, Woth K. 2007. Future extreme events in European climate: an exploration of regional climate model projections. Climatic
Change 81: 71–95. doi: 10.1007/s10584–006–9226-z.
Bruggeman V, Faure M, Haritz M. 2008. Schadensersatz für Opfer von Naturkatastrophen – Ein Vergleich zwischen Belgien und den
Niederlanden. Vierteljahrshefte zur Wirtschaftsforschung 77(4): 18–43.
Bouwer LM, Huitema D, Aerts JCJH. (2007). Adaptive flood management: The role of insurance and compensation in Europe. IVM (Int. rep.
W07/08). Amsterdam: Institute for Environmental Studies.
Buchanan JM. 1975. The Samaritan’s Dilemma. In Altruism, Morality and Economic Theory, Phelps ES (ed.). Russell Sage Foundation: New
York; 71–85.
Browne MJ, Hoyt RE. 2000. The demand for flood insurance: empirical evidence. Journal of Risk and Uncertainty 20(3): 291–306.
doi:10.1023/A:1007823631497.
CEA (European Insurance and Reinsurance Federation). 2005. The insurance of natural events on European markets. CEA, Brussels.
CEA (European Insurance and Reinsurance Federation). 2007. Reducing the social and economic impact of climate change and natural catastrophes.
Insurance solutions and public-private partnerships. CEA, Brussels.
Citlak B, Wagner GG. 2001. Hochwasser: Unkalkulierbare Ereignisse und unvermeidbare Folgen? Denkanstöße für den ökonomischen
Umgang mit Elementarschäden. In Private Versicherung und soziale Sicherung – Festschrift zum 60. Geburtstag von Roland Eisen, Mager
HC, Schäfer H, Schrüfer K (eds). Metropolis: 97–111.
Coate S. 1995. Altruism, the Samaritan’s Dilemma, and government transfer policy. American Economic Review 85(1): 46–57.
Commission of the European Communities. 2007. Green Paper: Adapting to climate change in Europe – options for EU action. {SEC(2007) 849}
COM(2007) 354 final, Brussels: 29.6.2007.
EU (2008). EU Council Directive 92/49/EEC of 18 June 1992 on the coordination of laws, regulations and administrative provisions relating
to direct insurance other than life assurance and amending Directives 73/239/EEC and 88/357/EEC (third non-life insurance Directive).
At: http://ec.europa.eu/index_de.htm [accessed 11 August 2008].
Fischer M. 2004. Solidarität schafft Sicherheit. Die Kantonalen Gebäudeversicherungen als System. In Katastrophen und ihre Bewältigung,
Pfister C, Summermatter S (eds). Haupt Verlag: Berne; 203–216.
Fischer M. 2008. Der lange Weg zur integralen Elementarschadenversicherung und -vorbeugung in der Schweiz. Vierteljahrshefte zur
Wirtschaftsforschung 77(4): 98–103.
Gore A. 2006. An inconvenient truth. The planetary emergency of global warming und what we can do about it. Melcher Media, New York.
Huber M, Amodu T. 2006. United Kingdom. In Financial compensation for victims of catastrophes, Faure M, Hartlief T (eds). Springer:
New York; 261–302.
Huber M. 2008. Zur Versicherung von Elementarrisiken. Das englische Gentlemen’s Agreement und seine Entwicklungsmöglichkeiten.
Vierteljahrshefte zur Wirtschaftsforschung 77(4): 44–52.
IPCC. 2007. Climate Change 2007: Impacts, adaptation, and vulnerability. Contribution of Working Group II to the Third Assessment Report of the
Intergovernmental Panel on Climate Change. Cambridge University Press: Cambridge.
Kirchgässner C. 2007. On the efficiency of a public insurance monopoly: The case of housing insurance in Switzerland. In: Public Economics
and Public Choice Contributions in Honor of Charles B, Blankart, Baake P, Borck R (eds). Springer: Berlin; 221–242.
Kron A. 2009. Mikroskalige Ermittlung potenzieller Hochwasserschäden zur Gefahren- und Risikoanalyse. Dissertation am Institut für Wasser und
Gewässerentwicklung Universität Karlsruhe (TH). Published as: Mitteilungen des Instituts für Wasser und Gewässerentwicklung -Bereich
Wasserwirtschaft und Kulturtechnik 236, Karlsruhe.
Kundzewicz WZ, Szamalek K, Kowalczak P. 1999. The great flood of 1997 in Poland. Hydrological Sciences 44(6): 855–870.
Kunreuther H, Linnerooth-Bayer J. 2003. The financial management of catastrophic flood risks in emerging-economy countries. Risk Analysis
23(3): 627–639. doi: 10.1111/1539–6924.00342.
Lasut A. 2005. Insurance as a Means of Reducing Economic Losses Due to Floods. In Non-structural Methods of Flood Prevention: Possibilities
and Constraints, Ignar S. SGGW: Warszawa; 75–85.
Michel-Kerjan E. 2001. Insurance against natural disasters: Do the French have the answer? Strengths and limitations. Working Paper of the Ecole
Polytechnique, Cahier No. 2001–007.
Ostrowski J, Dobrowolski A, Zaniewska M. 2002. The analysis of disastrous floods in Poland after second world war as a base of a new flood
control strategy. 5th International Conference on Hydro -Science & -Engineering. Warsaw, Poland.
DOI: 10.1002/eet
Pfurtscheller C, Schwarze R. 2008. What can save us from catastrophic loss? A multivariate analysis of the factors of socio-economic vulner-
ability against floods in Tyrol / Austria. Proceedings of the IDRC 2008. Davos.
Prettenthaler F, Vetters N. 2005. Extreme Wetterereignisse: Nationale Risikoausgleichssysteme im Vergleich. In Extreme Wetterereignisse und
ihre wirtschaftlichen Folgen, Steininger KW et al. (eds). Springer: Berlin; 91–114.
Raschky PA, Weck-Hannemann H. 2007. Charity hazard – A real hazard to natural disaster insurance? Environmental Hazards 7(4): 321–329.
doi: 10.1016/j.envhaz.2007.09.002
Raschky PA, Schwindt M, Schwarze R, Weck-Hannemann H. 2008. Risikotransfersysteme für Naturkatastrophen in Deutschland, Österreich
und der Schweiz – ein theoretischer und empirischer Vergleich. Vierteljahrshefte zur Wirtschaftsforschung 77(4): 53–68.
Raschky PA, Schwindt M, Schwarze R, Zahn F. 2010. Uncertainty of governmental relief and the crowding out of insurance. Working Papers in
Economics and Statistics 2010–03. University of Innsbruck.
Schwarze R, Wagner GG. 2004. In the aftermath of Dresden – New directions in German flood insurance. Geneva Papers of Risk and Insur-
ance 29(2): 154–169.
Schwarze R, Wagner GG. 2007. The political economy of natural disaster insurance: lessons from the failure of a proposed compulsory insur-
ance scheme in Germany. European Environment 17: 403–415.
SVV (Schweizerischer Versicherungsverband). 2006. Höhere Deckung bei Naturkatastrophen. Positionen der Versicherungswirtschaft 3.
Thenews.pl (21 May 2010). PM – flood Poland’s worst natural disaster. http://www.thenews.pl/national/artykul132062_pm—flood-polands-worst-
natural-disaster.html [accessed 2 July 2010].
Ungern-Sternberg T. 2001. Die Vorteile des Staatsmonopols in der Gebäudeversicherung: Erfahrungen aus Deutschland und der Schweiz.
Perspektiven der Wirtschaftspolitik 2(1): 31–33. doi: 10.1111/1468–2516.00030
Ungern-Sternberg T. 2002. Gebäudeversicherung in Europa. Die Grenzen des Wettbewerbs. Berne: Haupt.
Url T, Sinabell F. 2008. Flood risk exposure in Austria – options for bearing risk efficiently. Schmollers Jahrbuch (Journal of Applied Social
Science Studies), SJ 128: 593–614.
Zweifel P, Eisen R. 2003. Versicherungsökonomie, 2nd edn. Springer: Berlin.
DOI: 10.1002/eet

Natural Hazard Insurance in Europe - Tailored Responses To Climate Change Are Needed - Schwarze (2010)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Natural Hazard Insurance in Europe - Tailored Responses To Climate Change Are Needed - Schwarze (2010)

Uploaded by

Copyright:

Available Formats

Environmental Policy and Governance

Env. Pol. Gov. 21, 14–30 (2011)

Natural Hazard Insurance in Europe: Tailored

Received 7 November 2009; revised 30 July 2010; accepted 20 August 2010

Figure 1. Stylized models of natural hazard insurance

Stylized Models for Natural Hazard Insurance

Model 1: Public Monopoly Insurer

Model 2: Compulsory Insurance

Model 3: Coupling of Contracts

Model 5: Tax-ﬁnanced Governmental Relief Funds (Catastrophe Funds)

Insurance Systems Against Natural Hazards in Europe: an Overview

Regional Responses to Climate Change are Needed

Country Description Model

Country Description Model

Table 1. Natural hazard insurance in Europe

Figure 2. Sum of precipitation (l/m2)

Tyrol Upper Bavaria Grisons

State measures 53.3* 22.0* 100†

Table 2. Damage coverage (%)

Tyrol Upper Bavaria Grisons

Table 3. Time lag (days)

Tyrol Upper Bavaria Grisons

>1* >1† 0.3†

Table 4. Premiums (‰)

Case Study Résumé

You might also like