Is global warming insurable? A comparative analysis

Contents

Index of Abbreviations

Introduction

I. Insurability of natural hazards

II. Current institutional arrangements (USA, UK, Germany)

The United States
1. Floods in the United States
1.1 Main institutions involved with US floods
1.1.1 The NFIP
1.1.2 The SBA
1.1.3 Federal Government
1.2 Interaction between institutions
2. Windstorms in the United States
2.1 Private insurance
2.2 The Federal Emergency Management Agency
2.3 Governmental insurance regulation
2.4 Interaction between Institutions
2.5 Hurricane Katrina
3. Droughts in the United States
3.1 The perception of drought in the US
3.2 Main institutions involved with drought
3.2.1 Monitoring U.S droughts
3.2.2 The role of the RMA and FSA
The United Kingdom
4. Floods in the United Kingdom
4.1 Institutions managing floods in the UK
4.1.1 DEFRA
4.1.2 The Environment Agency
4.1.3 Private Insurance
4.2 Interaction between Institutions
4.3 The 2007 Floods
5.0 Storms in the UK
5.1 Institutions managing storms in the UK
5.1.1 Private Insurance
5.1.2 The Met Office
5.1.3 Central Government
5.2 Interaction between institutions
5.3 Impact of windstorms
5.4 Hurricane Kyrill
6. Drought in the UK
6.1 Institutions managing drought in the UK
6.1.1 The Environment Agency (EA)
6.1.2 Defra (Department of the Environment Food and Rural Affairs)
6.1.3 The water industry
6.1.4 Ofwat (The Water Services Regulation Authority)
6.1.5 Secondary institutions involved with drought
6.2 Interaction between institutions
6.3 Impact of drought
Germany
7. Floods in Germany
7.1 Institutions managing flood risk
7.1.1 German flood insurance
7.1.2 Federal Government
7.1.3 The role of State Governments
7.2 Interaction between institutions
7.3 Impact of floods
7.4 The Century-Flood “Jahrhundertflut”
7.5. Mandatory insurance in Germany
7.6 Flood effects in detail
8. Windstorms in Germany
8.1 Institutions managing windstorms
8.1.1 Private insurance
8.1.2 Central Government
8.1.3 Monitoring windstorms
8.2 Interaction between institutions
8.3 Impact of windstorms
8.6 Winterstorm Kyrill
9. Drought in Germany
9.1 Private insurance
9.2 Federal Government
9.3 Possible impacts of drought
9.4 The 2003 summer drought

III. Comparative approach
10. The United States
10.1 Floods
10.2 Storms
10.3 Droughts
11. United Kingdom
11.1 Floods
11.2 Windstorms
11.3 Drought
12. Germany
12.1 Floods
12.2 Storms
12.3 Droughts
13. Functional difference between institutional arrangements

IV. Global warming a perpetuation
14. The United States
14.1 Are floods in the United States insurable?
14.2 Are windstorms in the United States insurable?
14.3 Insurability of United States droughts
15. United Kingdom
15.1 Are floods in the United Kingdom insurable?
15.2 Insurability of windstorms in the UK
15.3 Insurability of drought in the UK
16. Germany
16.1 Insurability of floods in Germany
16.2 Insurability of windstorms
16.3 Insurability of droughts in Germany
17. Insuring natural hazards in a global warming environment
17.1 The protective institutional arrangement
17.2 National differences and prospects
17.3 Insurability of natural hazards

Conclusion

References

Tables

Figures

Index of Abbreviations

illustration not visible in this excerpt

Introduction

The institution of insurance as it is known today has been largely neglected by sociologists, even though it is the central institution, alongside government, responsible for risk management (Ericson et al 2003: 44). A central focus of risk management is the protection of society from damage caused by natural hazards. This thesis aims to frame insurance and other institutions involved with risk management into a particular viewpoint of social problems.

Natural hazards constitute the largest shore of the market for insurance while at the same time pose the largest challenge for the institution. One may question why this is so. Well into the 21st century anthropogenic climate change or global warming is no longer merely a theory, it has become reality. International reinsurers, scientists and politicians agree that anthropogenic CO2 and other gases are increasing the temperature on earth. Consequentially natural hazards are increasing in frequency and impact (Munich Re 2009). This poses problems for societies dealing with impacts from natural hazards. Much research has been done on individual behavior in deciding whether or not to insure against natural hazards (Kunreuther 2004; Kerjan 2008a; Grace 2006). This line of research views individual as myopic in assessing risk, thereby underestimating risk and the relevance of insurance. The emphasis of this research is mostly on the irrationality of individuals as regards the benefits of insurance. This perspective fails to take into account the influence of societal institutions on individual decisions; for example, whether or not to insure against disasters. Theoretical exercises in this direction are common, whereas attempts to consider the environment of a specific risk management system are rare. Michael Huber’s observation (2004a) represents an exception as it describes the institutional arrangement for the management of floods in England while emphasizing the influence it has on insurability.

The assumption that institutions exercise an effect on the insurability of natural hazards remains central. Generally when regarding natural hazards one can differentiate between those affected and those unaffected by anthropogenic climate change. Natural hazards which are not amplified by global warming are earthquakes, volcano eruptions and tsunamis (Munich Re 2009). In turn floods, windstorms and droughts will increase in impact and frequency due to anthropogenic climate change; hence they will be the central theme in the following. Institutional arrangements for the management of weather related disasters emerge as interesting as these will be most adversely affected by accumulating events. The underlying question is: are floods, storms and droughts insurable under current institutional arrangements? This question is posed for the USA, UK and Germany and answered in this thesis over four sequential parts. Part I explores the history of the institution of insurance and its development to a concept applicable for various social uncertainties. Arising from this natural hazards are discussed as a paradoxical phenomenon – while on one hand the market in which insurers offer coverage is huge and profits are large, on the other hand the insurability of natural hazards is starkly questioned. The role of government in assisting insurers is tackling the central challenge of the business of insurance - adverse selection. Part II sets out the institutional arrangements for management of floods, storms and droughts, highlighting typical behaviour of politicians and insurers. Particular events or changes in policies are studied with the intention of creating a complete picture of risk management. Interactions between institutions are investigated and commented on. Arising from this a functional equivalence for systems managing natural hazards is assumed (Part III). This is tested by observing institutional arrangements in respect of their stability in the face of systemic disturbances. Elements which prevent a system from functioning optimally are identified. The varying extent of governmental involvement among the countries analyzed is discussed while the influence this may have on the insurability of weather related disasters is considered. Based on knowledge of differences between countries’ risk management regimes their methods of dealing with an equal threat are investigated. Part IV assesses global warming as a permanent factor, creating new risks for all involved. The institutional arrangements for the management of natural hazards are considered with respect to a potential increase in the number and severity of impacts. More specifically it is suggested whether or not systems will prevail in their particular environments. The behavior of individuals is considered with regard to how they base their decisions on whether or not to purchase coverage, taking into account the presence of surrounding institutions.

Emerging from these thoughts is a concept to determine the influence of an institutional arrangement on the insurability of a natural hazard. The following analysis aims to frame various dimensions of risk into an understandable picture of natural hazard management.

I. Insurability of natural hazards

Social scientists have seldom investigated insurance, except for some narrow specialisms of law and economics. Concerning sociology, Ericson et al. (2003) state that “… the sociology of insurance remains nascent”. They follow this by offering a theory of “Insurance as Governance” which illustrates how insurance, alongside central government, is the main contributor to social security, solidarity and freedom (Ericson et al: 2003). Other sociological contributions view insurance under different perspectives (Burchell 1991; Beck 1999; Stone 1999; Huber 2008).

Generally the concept of insurance as such can be tracked back as far as the first millennium. The procedure of insuring their freight against losses remained feasible for seamen aiming to meet the expectations of an investor throughout the middle age (cf. Luhmann 1993: 17) (cf. Jaffee et al. 1999: 207). The range of populations interested in this concept hardly changed over the following centuries. The institution of insurance, as known today, arose at the beginning of the 19th century. It evolved from a combination of growing wealth and society’s demand to be able to deal with risk. The coverage of industrial labour accidents was the first example of a successful application of insurance. At first opposed by the workers of the 19th century, compensation for industrial labour accidents developed to be an appreciated element of modern society (Burchell 1991: 197). This was the introduction of insurance into modern society. Following this, a tendency grew for governments to promote the benefits connected with the purchase of private insurance policies. For example, by the end of the nineteenth century governments encouraged individuals to purchase life insurance policies, glad to be rid of the responsibility taking care of surviving dependents (Ericson et al. 2003). Rapidly society accepted and embraced these new usages of the institution. Insurance evolved into a concept that could be applied for dealing with various social uncertainties. At the same time, insurance evolved to an institution of social control sharing the goals of “providing security and solidarity through the pooling of risks” with the state (Ericson et al 2003: 47). In this sense insurance is a social instrument as long as individuals pay premiums and agree to rules of participation (cf. Ericson et al 2003: 49). Along with growing scientific and general knowledge of risks, insurance became an institution which was relied on more and more. Emerging as “technology of risk”, insurance combines multiple known facts about society into a specific set of rules (Burchell 1991: 197). These rules are not static but vary with institutional and environmental constraints. One can, however, generalize five characteristics that constitute insurance’s rationality, enabling it to be efficient:

I. The ability to spread risk
II. The capacity to reduce the variance of risk
III. The segregation of risk
IV. The encouragement of loss reduction
V. The ability to monitor and control the behaviour of the insured. (cf. Freeman 2007: 165)

Among insurance scholars the “law of large numbers” is the core reason why the institution of insurance functions. “In simple terms, this law states that, as the number of independent events (here, losses) under consideration increases, the frequency distribution of those events tends toward the normal distribution. Therefore, the variance of the mean value decreases as the number of events increases. If we view the variance as a measure of volatility or risk of loss, then the risk associated with the loss reduces as the number of policies grows (Freeman 2007: 165)”. Other reasons for its functioning are sharing risk with other insurers (I), diversifying risk through large numbers of policies (II) and segregating policy costs by introducing high and low risk categories for the insured (III). Occasionally incentives are introduced to lower the risk exposure of the insured (e.g. health insurers may financially reward non-smokers) (IV). Finally, insurers monitor the insured, mainly to verify that they operate in a manner consistent with underwriting standards (V) (Freeman 2007: 167).

Once these typical characteristics had been identified the institution of insurance continued using the method of classifying events as risks. The birth of social insurance near the end of the nineteenth century was such a development (cf. Burchell 1991: 197). Insurance’s ability to adapt to new constraints and environmental effects has been described as highly successful (cf. Burchell 1991: 198; Crichton 2002: 239 Freeman 2007: 165). Despite this, a new challenge for the insurance industry emerged during the 20th century. Natural hazards had been striking the earth long before man first set foot on it. However changes had been occurring during the 20th century. Firstly, society’s accumulation of wealth resulted in a demand for coverage against natural hazards. Secondly, an increasing population settled in hazard prone areas. Thirdly, anthropogenic climate change had become reality enhancing the frequency and impact of natural hazards (cf. Munich Re 2009). These changes caused problems for insurance. Diversifying and segregating risk is a difficult challenge when large populations settle in hazardous areas. Beyond these societal changes, natural hazards also pose insurance problems because of their specific character of rarely occurring while impacting immensely. This leads to several intricacies. These catastrophes are known in the insurance literature as low probability high consequence (LPHC) events (Grace et al. 2001: 8). Widespread in industrial nations, the natural hazard insurance market is meanwhile the largest worldwide. Economic losses from natural hazards have risen from $53.6 billion (1950–1959) to $620.6 billion (2000-2008) (Kunreuther 2009: 4). Insured losses have risen by $3 to $4 billion annually in the 1970s to record annual sums of up to $87 billion in 2005^[1]. Kunreuther (2009) states that, financing natural hazards challenges insurers and the political agendas of many countries. Huber (2008) points out two contradicting views on the insurability of natural hazards from a sociological perspective. On the one side Richard Ericson et al. (2003) “…assumed total insurability due to the universal applicability of risk, only somewhat constrained by economic factors (Huber 2008: 2)”. Whereas on the other side Ulrich Beck “…emphasised the weakening of the methodological fundament of risk management due to doubt, lack of legitimacy and an eroded scientific rationality” and he concludes a decline in insurability (Huber 2008: 2). Whichever view is more accurate, since the turn of the millennium the insurability of natural hazards is frequently being questioned. Assuming they are not insurable, what then determines an uninsurable risk? Jaffee et al. (1997) state three factors as impediments constituting an uninsurable risk: adverse selection, moral hazard and that the risk is too large for precise calculation. With respect to natural hazards the problem of moral hazard can be controlled by surveillance and audit; moreover the fact that disaster impacts may be too large has in the past never led insurers to deny coverage (Jaffee et al. 1997: 206). They continue to point out that disaster risk is not private information of the insured and that the insured has no control over the event for which coverage is provided (Jaffee et al. 1997: 206). Somewhat contrary to this optimistic view, it is assumed in the following that adverse selection is the central challenge when considering the insurability of natural hazards. The main reason is that the proportion of society vulnerable to natural hazards is often located within a particular area. This troubles insurers, as for insurance to be efficient it requires the segregation of risks over several areas and over different classes of insured. Considering floods and windstorms it is evident that these more often impact on coastal areas than elsewhere. Therefore homeowners settling in these areas are more likely to purchase coverage from floods and windstorms than homeowners residing in “safer” areas. This is the point in which natural hazard insurance differs from other policies. The law of large numbers fails it function if a majority of the population insured is located within hazard prone areas. In this case even if the number of policy-holders were infinitely large, financial losses would still outweigh the profits. Under normal conditions insurance functions by mixing good with bad risks to an even proportion; referring to life insurance this would mean insuring healthy young individuals and sickly old individuals in equal numbers. The twist with natural hazards is that the majority of individuals purchasing coverage are aware that they are exposed to risk, whereas those not purchasing coverage are also aware that they are residing outside of threatened areas. Consequentially insurers of natural hazards are often burdened with too many bad risks. The problem for an insurer with a majority of bad risks is commonly referred to as adverse selection^[2]. Scott E. Harrington states that a precondition for adverse selection is asymmetric information between insurer and insured; “insurers must be unable to identify high-risk buyers. Low-risk parties may rationally not insure (Harrington 1999: 41)”. A further assumption is that a more accurate risk classification of purchasers of insurance lessens the problem of adverse selection (cf. Harrington 1999: 41). These assumptions may be imprecise; regarding natural hazards insurers are most often able to identify high-risk buyers but are left powerless to suffer from adverse selection if high-risk buyers are the only ones willing to purchase insurance. Equally if a population requesting coverage is dominated by high risk individuals, then the advantage of classifying in groups diminishes.

Coping with adverse selection is the central hurdle when considering the insurability of natural hazards. If insurers struggle to include a diversity of risk classes, then paying indemnification will most likely fail. In essence the only solution to this problem is to redistribute resources; either the many contribute to the few, or the many participate by paying into an insurance scheme. Whichever way an approach is attempted the government must act as an intermediate. If this is the case then the state is left with three possibilities to integrate the proportion of society unaffected by natural hazards. Firstly, it can introduce mandatory insurance for all weather related damages, thereby securing a wide range of participation. This may be bundled with other risks or as a single policy. In both options society is obliged to invest in insurance. Secondly, it can subsidize insurance policies hence creating an artificial situation in which low risk groups see the benefit of purchasing coverage. In this situation society pays for subsidies in form of general taxes and may additionally voluntarily purchase coverage. Thirdly, the state can act as a reinsurer covering losses from natural hazards, thereby forcing taxpayers to finance deficits in the insurance industry. The first two options leave insurers with a composition of good and bad risks. The third option merely drains tax money. In this sense only if government can succeed in establishing an even share of good and bad risks, may government intervention be considered an effective tool to assist the industry in insuring natural hazards. Notably, even if governments succeed in diversifying risks, the problem of adverse selection has not been solved but merely transferred to another dimension^[3]. The price for settling in hazard prone areas has to be paid, whether by society as a whole or by those exposing themselves to risk. However, currently the insurability of weather related disasters remains an open question. The main topics are whether or not natural hazards are insurable, and if so then how? In this context the leader in reinsurance Swiss Re, a contributor to the discussion, stated in 1998 that the insurability of floods is decided by six factors:

I. Mutuality: A large number of people who are at risk must combine to form a risk community.
II. Need: When the anticipated event occurs, it must place the insured in a situation of financial need.
III. Assessibility: The expected loss burden must be assessable.
IV. Randomness: The time at which the insured event occurs must not be predictable and the occurrence itself must be independent of the will of the insured.
V. Economic viability : The community organized by the insured persons must be able to cover its future, loss-related financial needs on a planned basis.
VI. Similarity of threat: The insured community must be exposed to the same threat and the occurrence of the anticipated event must give rise to the need for funds in the same way for all concerned. (Swiss Re 1998: 7)

Swiss Re (1998) states, regarding risks from natural hazards, that “Mutuality”, “Assessibility” and “Economic Viability” pose the largest obstacles to insurability. These constraints were applied for floods; however they are also valid for other natural hazards. Populations living in coastal areas are more vulnerable to windstorms than in non-coastal areas. Droughts affect the agricultural sector more excessively than the remaining population. This suggests that the insurability of storms and droughts is defined by similar constraints. Referring to mutuality, large vulnerable populations are merely found on a global scale. As natural hazards strike irregularly, impacts are often unforeseeable and vary largely in severity. This poses difficulties for insurers aiming to create sizable risk communities. Huber (2008) stresses the factor of “Economic Viability”, stating that societies must be economically wealthy to insure against rarely occurring events. However, even in prosperous areas individuals’ willingness to invest in coverage is limited, despite their being exposed to risk. In respect of floods, less than 1 per cent of an average country’s real-estate stock is seriously exposed. So the risk community is too small to ensure an economical solution for insurers and insured (Swiss Re 1998: 7). On top of the size and wealth of exposed populations the largest problem remains the fact that individuals willing to insure against natural hazards tend to be aware of their vulnerability^[4]. Likewise individuals residing outside of risk areas are uninterested in insurance policies. Abstracting from the above, there are three main factors opposing the insurability of natural hazards:

I. The steep rise in financial costs due to impacts from natural hazards
II. The requirement to form risk communities, financially able and willing to insure against natural hazards
III. The challenge to insure a relative share of good and bad risks

Once these impediments are overcome, natural hazards can be considered insurable. Whether this challenge should be taken on by institutions (governments and/or insurance) or left to individual responsibility is discussed in the following.

Binding insurability to institutions

Scholars differ on the question as to whether natural disasters should be insured autonomously by private insurance companies or in combination with local government. Kovac (2001: 6) stresses the relevance of political and individual responsibility. In his view individuals should, independently from external incentives, inform themselves about hazard risks. Also governments may well benefit from introducing land-use measures, to discourage populations from settling in hazard prone areas. In addition governments ought to enforce building codes and preferably invest in structural defenses. In turn, Howard Kunreuther emphasizes the importance of public-private partnerships (PPP´s) for disaster management, focusing on regulations and standards (building codes) aimed at supporting the insurance industry (Kunreuther 2001: 3). Other opinions state that governmental policies undermine risk management (Kleindorfer 1999: 1). These policies (subsidies, rate-caps) remove incentives to manage catastrophe risk appropriately. The latter suggests that governments may interfere with insurance companies’ management of natural hazards. Government is only one of several institutions interfering with insurance. We can step back from the preceding perspectives and highlight the fact that all institutions involved are having major effects on the insurability of natural hazards. For example, the availability of disaster aid or subsidized insurance policies may bias an individual considering settling in a hazardous area. One may also expect a person, seeing the existence of institutional support mechanisms, to alter the extent of risk to which he will expose himself. What is more, an individual socialised within such institutions may be even unaware that he is shaping his decisions based on their presence. This suggests that institutions managing natural hazards impose an unknown influence on individuals. In this thesis the assumption is central that institutions influence individual behaviour when considering whether or not to insure against natural hazards. In this sense it is relevant to remark that government’s degree of influence varies among countries. Globally countries’ systems vary between no and complete governmental management of natural hazards. There are four categories within which nation states tend to manage natural hazards:

I. Governments autonomously manages natural hazards (USA)
II. Insurance is the only institution managing natural hazards (UK, Ireland, Israel)
III. Natural hazards are managed by a combination of federal government and insurance industry (France, Germany, Hungary, Italy)
IV. Individuals are fully exposed to natural hazards as no institution manages natural hazards (most Third World countries)

The question arises - are these solutions functionally equivalent? The fourth category (IV) is without function as it ignores the consequences of hazardous impacts. Of the remaining three one might assume it does not matter whether the institution of insurance, the government or a joint effort between the two manages risk. However, this needs to be put to the test. Is the management of natural hazards by insurance, government or a combination of both functionally equivalent? This thesis is more thoroughly pursued in the following chapters. More precisely, countries (of the first three categories) are analyzed concerning their institutions and local peculiarities (Part II). Then these institutional arrangements managing natural hazards in the United States, the United Kingdom and Germany are compared (Part III), thereby allowing assumptions as to whether or not systems managing natural hazards are functionally equivalent. A relevant factor in this context is a system’s resistance to disturbance . Conceivable as disturbances are: an extraordinary large hazardous impact; a series of extremely destructive hazards; the introduction of binding regulatory constraints; a sudden increase of individuals settling in hazardous areas.^[5] In this sense institutional arrangements are considered functionally equivalent if they are equally resistant to systemic disturbances. In turn, if institutional arrangements’ resistance to disturbances varies starkly, this will allow assumptions on their stability in respect to managing natural hazards.

Institutional arrangements

Within this thesis several countries and their institutions are considered in respect to natural hazards. Whenever natural hazards are mentioned within this thesis I refer to those enhanced by anthropogenic climate change: floods, windstorms and droughts. Government itself, governmental institutions and private insurance are often central elements managing natural hazards. Regardless of the relevance a particular institution engages, a combination of one or more is in the following referred to as an institutional arrangement. These institutional arrangements may differ on two dimensions. Firstly, countries vary on the extent they rely on government or private insurance for risk management. Secondly, institutional arrangements differ on their range of influence. Whereas some countries choose to manage natural hazards with a single institutional arrangement other countries have three autonomously running arrangements^[6]. This fact impedes efforts to compare a single natural hazard over several countries. Therefore, this thesis is constructed as a cross comparison, analyzing floods, storms and droughts in three different countries. Risk management is compared between natural hazards and these again over countries, aiming to offer a wide perspective on possible institutional arrangements. Globally society differs on various dimensions: environmental, social and political – the following analysis aims to frame these dimensions inside an understandable picture of natural hazard management.

II. Current institutional arrangements (USA, UK, Germany)

The United States

The United States’ governance of natural hazards is characterized by a governmental insurance scheme. Insurers merely write policies; however, this is not the case for windstorms as these are insured through private insurance. Additionally federal government frequently issues disaster aid for victims of disasters. Individuals are left to choose between insurance, disaster aid and the chance that fate will leave their property undamaged by natural hazards.

1. Floods in the United States

The previous decade has seen an increase in the number of major flood incidents. Both government and individual failings are discussed in relation to financial losses flood damage. The following chapter describes primary institutions dealing with flood damage in the United States (1.1.). The functions of these institutions are intended to integrate, thereby creating a sustainable system of flood risk management. The intended interactions between institutions will be discussed (1.2.).

1.1 Main institutions involved with US floods

Federal government is the dominant institution managing flood prevention and damage in the United States. It charges several institutions with monitoring risk and limiting disaster impacts. The central institution managing flood risk is the National Flood Insurance Program (NFIP) - its function and its history is described (1.1.1.). The need to supply catastrophe victims with financial aid to rebuild their houses is covered by the Small Business Administration (SBA) which was commissioned to allocate disaster-credits. The basic requirements for disaster aid and the SBA´s role in risk management are described (1.1.2.). In recent years flood incidents have been striking areas which were formerly assumed to be unaffected; this has led to large parts of society suffering flood damage without previously purchasing coverage. For this reason federal government has issued extensive financial compensation. The procedure of federal compensation and the costs arising from this are examined (1.1.3.).

1.1.1 The NFIP

In the United States, there is no private insurance scheme that provides coverage for flooding. Homeowners’ insurance policies explicitly exclude coverage for flood damage (Kerjan, 2008: 4). The main provider for flood coverage is the NFIP, founded in 1968, with the main function of giving homeowners access to federally backed insurance. The NFIP running under FEMA^[7] defines Special Flood Hazard Areas, so called SFHAs, in which homeowners with a federally backed mortgage are required to purchase insurance. This was enforced through the Flood Disaster Protection Act of 1973, prohibiting federal lending agencies from providing homeowners with loans of any kind if they had not previously purchased flood insurance (Swartz et al. 2006: 16). More generally the Act of 1973 had four main goals:

I. Substantially to increase the limits of coverage authorized under the NFIP
II. To provide accurate available information concerning flood-prone areas
III. To litigate federal financial assistance, for states and local communities to participate in the NFIP
IV. To litigate flood insurance to all property owners in flood prone areas receiving federal assistance of any kind

The American Congress summed this up as “to expand the national flood insurance program by substantially increasing limits of coverage and total amount of insurance authorized to be outstanding and by requiring known flood-prone communities to participate in the program and for other purposes^[8] ”. In 2008 the Flood Insurance Reform and Modernization Act was passed following several difficulties of the NFIP. Its main concerns were^[9]:

I. To exclude certain properties from receiving subsidized premium rates ( Non primary home-owners, those previously receiving more benefits than market value of property, house-owners refusing to accept any offer for mitigation assistance by government)
II. Introduction of mandatory insurance, for all residents of designated flood areas behind dams and other man-made structures.
III. Repeal of the necessity for FEMA director to consult other institutions on premium rates. Only condition these must respect current principles and the NFIP´s historic dept must be covered
IV. Suggests adding building codes to the NFIP act of 1968

Additionally, the Act amplified the requirement of Flood-Insurance for mortgage. It also included an increase of penalties for federally backed lenders who did not enforce mandatory flood insurance policies for homeowners in Flood-Zones (Kerjan, 2008: 38). Congress main comments on the Act are the following: The interest on the FEMA´s $20 Trillion borrowed for the 2005 hurricane season exceeds $1 trillion which FEMA will remain unable to repay. Flood risk outside 100-year floodplains is substantial, structural defences are not available nor are residents aware of potential risks to their lives or property^[10].

Leaving aside criticism on current procedures it is worth mentioning that FEMA runs a Floodplain Management Program. This Floodplain Management Program contains requirement-standards for new constructions. These requirement-standards include several forms of mitigation procedures for floods. “Currently over 20,100 communities voluntarily adopt and enforce local floodplain management ordinances that provide flood loss reduction building standards for new and in development buildings^[11].” In and around floodplain areas the number of policies has increased since inauguration of the NFIP. The amount of policies signed through the NFIP has risen from 1.5 million in 1978 to near six million in 2009^[12]. Comparing the amount of policies with the actual value of coverage illustrates that a few vulnerable states account for the majority of the exposure. Currently the state of Florida accounts for 40 per cent of nationwide coverage, exposing $454 billion of assets to flood risk. In 2007 premiums collected in Florida amounted to $900 million, whereas nationally a total of $2.81 billion in premiums was collected. Evidently, a minority of US states comprise the largest share of exposure insured by the NFIP. For example - the top ten states represent $947 billion of flood coverage or 85 per cent of the total value insured by the NFIP (Kerjan 2008: 12). The average premium collected nationally per policy is $506. Whereas the hazard prone state of Florida at $411 per policy suggests high state-subsidies, for comparison premiums in New York average at $757, almost double the cost (cf. Kerjan 2008: 12).

Returning to the steep rise of coverage provided through the NFIP it seems problems are unavoidable. Between 1978 and 2009 total coverage has risen from $50 billion to $1.2 trillion; while losses due to the 2005 hurricane season amounted to nearly $18 billion^[13]. Especially as a result the program’s future is severely questioned.

illustration not visible in this excerpt

Figure 1: Coverage allocated by the NFIP for the period 1978 to 2006

Six million policies amount to half of the properties located in flood-endangered areas, leaving the remaining half uninsured and fully exposed to flood events. Studies on market penetration assume that forty-nine per cent of eligible properties inside the SFHAs and around one per cent outside of SFHAs are insured through the NFIP (Kerjan, 2008: 5). Properties inside SFHAs face a higher probability of being hit by a significant flood than properties outside. A significant flood is determined by the number of losses paid. “A significant flood in the US is defined as a flood with more than 1.500 paid losses, since 1978 there have been 93 significant floods^[14].” Between 1953 and 2009 there were 1.191 declared major flood disasters^[15]. In the case of a declared flood disaster federal assistance is allocated for victims. (This is discussed at length in chapter two.)

1.1.2 The SBA

The Small Business Administration (SBA) issues financial disaster loans in the United States. A common procedure for flood victims lacking financial resources to pay for structural damage is to apply for a disaster loan through the SBA. The SBA was founded in 1953 and its main purpose is to secure and assist new, small and growing businesses. However it also provides loans for homeowners and businesses affected by natural disasters.

“SBA provides low interest disaster loans to homeowners, renters, businesses of all sizes and private, non-profit organizations to repair or replace real estate, personal property, machinery & equipment, inventory and business assets that have been damaged or destroyed in a declared disaster^[16].” The SBA also provides Mitigation Loans, eligible for homeowners pre-approved for disaster loans. The Mitigation Loan is issued in addition to the approved loan but may not exceed 20 per cent of the original disaster loan^[17]. The SBA disaster loans are the primary form of federal disaster assistance given for rebuilding and repairing private structures. By its own account the SBA provides affordable, timely and accessible financial assistance to homeowners through its Office of Disaster Assistance (ODA)^[18]. However, with the increasing intensity of natural disasters in the United States, the SBA has been struggling to cope with the amount of applications. After Hurricane Katrina the 422,000 requested disaster loans raised the target of 21 days’ process time to 74 days, which posed a problem that could partly be solved through the Accelerated Disaster Response Initiative (ADRI) ( Kunreuther et al. 2007b: 8). The SBA can be seen as an effective support to the financial sector in issuing credit after natural hazards.

1.1.3 Federal Government

The United States Government plays a central role in the management of flood disasters. Before the inauguration of the NFIP in 1968 insurers had refused to include coverage for flood incidents in standard property insurance policies (Burby 2006: 173). Since 1968 there have been 1039 presidentially declared flood disasters^[19], followed mostly by disaster aid. Beyond these, several items of legislation have been passed on the subject of flooding and the mitigation of floods. The National Flood Insurance Reform Acts of 1994 and 2004 were followed by the most recent Flood Insurance Reform and Modernization Act of 2008 (mentioned above). Adjacent to the latter on the 30th October 2000, the United States president passed the Disaster Mitigation Act (DMA)^[20]. This act amended the Robert T. Stafford Disaster Relief and Emergency Assistance Act which was signed into law 28th March 1988^[21]. The DMA emphasises the necessity of both state and local mitigation efforts, thus mandating state mitigation plans as a requirement for disaster assistance. Since their inauguration in 2000, state mitigation plans are a precondition for post-disaster assistance. Additional funds were allocated for development of state and local mitigation plans^[22]. Returning to the procedure of managing flood events by issuing disaster aid, this process is now briefly explained and more thoroughly discussed in chapter two. After a hazardous event has occurred the governor of an affected state may request assistance from federal Government^[23]. After a request, the President decides whether or not to declare a disaster^[24]. Only after a governor’s request and only if the president declares disaster can assistance be given (Garett et al. 2002: 2). Since 1953 the total number of flooding disasters declared in the US has been 1.193^[25]. The amount of disaster assistance distributed is decided by congress (Kunreuther et al. 2007: 7). Federal disaster assistance has developed into the most central form of dealing with flood incidents. The United States distributes, through its institutions (FEMA, SBA), an average of $9.5 billion disaster assistance annually (cf. Burby 2006: 174). A statement of the three central institutions concerned with flood management in the United States was given. There follows an illustration of the interaction between these institutions.

1.2 Interaction between institutions

The insurance policies written under the NFIP are nearly all distributed by private insurance companies (Rabin et al. 2006). Federal government issued a bill enabling this procedure while providing a connection between state, private insurance and homeowners.

This effort named “Write-Your-Own (WYO)” Program^[26] was passed through congress in 1983 allowing property and casualty insurance companies to sign policies in their own name. Insurers collect the premium, pay and defend all claims; the NFIP merely steps in when losses are involved. The insurers, - however, are required to charge the rate given by the NFIP which is risk based and reflects the conditions of the affected state (Rabin et al. 2006). The NFIP acts as a reinsurer for private insurance companies with the difference that in this case the private insurer carries no risk at all (Kerjan, 2008: 10). Rabin et al. (2006) state that without existence of the NFIP $11.9 billion in claims would have been paid by taxpayers through federal disaster assistance funding since 1973. Further flood plain management standards and mitigation programs have avoided $1 billion in damages every year (Rabin et al: 2006). This federally initiated connection between NFIP and private insurance companies is also supported through the credits allocated by the SBA. The SBA provides mitigation credits which can be used to secure buildings and therefore lower the premium paid for flood insurance. SBA credits are given at lower rates than banks would offer. However, these are only available if the individual was denied credit by other finance institutions. The NFIP also aims to influence state governments, by pressuring governors of flood-prone states to enforce mitigation laws, building codes and early warning systems to deal with future flooding. In theory, this would lower the possibility of losses through floods. The higher risk awareness encourages finance houses to provide credit for homeowners to mitigate. The increase of mitigation measures releases tension on the NFIP in its role to reinsure risks associated with flooding. The encouragement of mitigation also lowers the amount of high end ex-ante disaster assistance payments issued by federal government. This institutional setting aims to protect society from floods to provide aid for as many individuals as possible. Furthermore it transfers ex-ante disaster assistance into pre-costs for homeowners and business-owners. The intended result is to release flood unaffected taxpayers from costs.

2. Windstorms in the United States

Over the last decade windstorms have been a major problem for the American society, striking frequently and causing billions of dollars worth of damage. This chapter examines how institutions deal with windstorms within the United States. The focus is on the private insurance scheme as the primary institution managing storm risk (2.1.). Following disasters joint efforts coordinate the assistance of victims requiring aid. The main institution acting in this context is the Federal Emergency Management Agency (FEMA), which is thoroughly described in this chapter (2.2.). The discussion of insurance regulation has recently become more intense, as a result of the impacts of major hurricanes. Even when only in discussion, insurance regulation has been carried out for several decades by state government. An explanation of the basic elements of regulation is given and its effects are commented on (2.3.) Then interactions between institutions are discussed (2.4.). The chapter ends by highlighting the devastating impact Hurricane Katrina made on the American society in August 2005 (2.5).

2.1 Private insurance

The only coverage available for windstorms is provided by private insurance. As is typical with other risks such as fire, hail, riot, lightning, explosion and theft, US storms are insured in a so called “Homeowners multi-peril insurance package” (Grace et al. 2001: 8). These packages are restricted to residential structures. Insurers may exclude some of the aforementioned perils. Storms may be excluded as long as they are not bound to a mortgage which is mandatory for residents of coastal areas. Market concentration of insurers varies considerably. Major disasters often cause companies to leave a market. In the late nineties the number of insurance groups decreased while individual insurers accumulated (Grace et al. 2001: 18). An explanatory factor is that premiums tend to rise after major disasters. This attracts new firms. In 1998, the average annual premiums varied from $574 in New York to $944 in Florida; nationwide the average annual premium was $578 (Grace et al. 2001: 28). By 2001, the average premium for Florida had decreased to $673. Only few years later Hurricane Katrina caused Florida’s average premium levels to peak at $1,193, US average levels remaining slightly lower at $886 (Grace et al. 2006: 32). These premium variations pose a problem for insurers entering or exiting a state market.

To retain the availability of insurance hazard prone states tend to enforce high exit barriers. As an unintended consequence this lowers the number of insurance groups. Companies intending to enter a market are discouraged by high exit barriers. Insurers are also confronted with a more basic problem in insuring storms. Storms pose the difficulty of estimating the risk of a future occurrence. More basically natural hazards are known as low-probability high-consequence events called (LPHC) (Grace et al. 2001: 8). LPHC’s are characterized by a low probability of striking but with possible very high consequences. Additionally, there are a large number of variables connected to calculating the impacts (Grace et al. 2001: 14). The major factor of uncertainty is the amount of damage a hurricane will inflict. Whenever damage is higher than expected by insurance industries loss ratios will raise. In Florida after Hurricane Andrew for example, loss ratios soared to 990.3 per cent and caused the nationwide loss ratio to rise to 124.6 per cent for the year of 1992 (Grace et al. 2001: 31). This resulted in the insolvency of 12 insurers. The reasons were the unexpected strike of disaster and that firms had been charging premiums which were too low for preceding years (Kleindorfer et al. 2007: 13). Hurricane Katrina offered a similar picture in 2005, causing roughly $40 billion in insured losses representing the largest insurance loss in US history (Swartz et al. 2006: 9). When insurers fail to assess risk, companies become insolvent and are replaced. The combination of insolvent firms and higher premiums has an attracted other insurance companies to enter the market. A majority of losses is absorbed by the insurance sector. Costs not covered by the insurance sector may be uninsured damages and damage to federal property. The agency responsible for such matters is the Federal Emergency Mamagement Agency (FEMA).

2.2 The Federal Emergency Management Agency

FEMA, which claims that it can be traced back to a congressional act of 1803, coordinates the federal Government disaster management. “The Federal Emergency Management Agency coordinates the federal government's role in preparing for, preventing, mitigating the effects of, responding to, and recovering from all domestic disasters, whether natural or man-made, including acts of terror^[27] ”. In 1988 the Robert T. Stafford Disaster Relief and Emergency Assistance Act was passed allowing the US Government to release federal funds through FEMA (Faure et al. 2003: 315). “Federal assistance under the Stafford act should be awarded when the incident is of such severity and magnitude that effective response is beyond the capabilities of the state and the affected local governments and that federal assistance is necessary (Garett 2002: 4).” FEMA’s disaster assistance is restricted to an annual budget defined by congress. Every further dollar of financial aid must be passed through congress in the form of a bill. This was the case for major disasters like Hurricane Andrew and Katrina. A storm classified as a major disaster must reach a certain wind speed^[28]. Storms declared as disasters outreach local capabilities and are managed by FEMA. Since 1954 the United States Presidents have announced 292 disasters in form of hurricanes and tropical storms^[29].

The standard procedure of handling disaster is:

I. A governor requests federal aid
II. A Preliminary Damage Assessment (PDA)^[30] decides whether the disaster has surpassed state and local means
III. If this is the case, the President declares a disaster
IV. FEMA provides disaster aid. (Faure et al. 2003)

By FEMA´s own claims it aims to return a damaged region back into a functional state (Faure et al. 2003: 317). At first emergency support is seen to; after which loans for property damage are allocated for. A long term prospect is given by providing a hazard mitigation program^[31]. Notably, the maximum rate of disaster aid paid to an individual or household is capped at $25,000 (cf. Kunreuther et al 2006b: 2). If the disaster considered is a hurricane and is predictable, FEMA orders an Advance Emergency Response Team (ERT-A)^[32] to the endangered area. This team coordinates early response efforts and identifies what kind of further assistance will be needed (Faure et al. 2003: 317). FEMA has a broad spectrum of influence on all natural hazards. However, FEMA was discussed in this chapter due to its central role in managing hurricanes. There follows s a description of the regulation of insurance.

2.3 Governmental insurance regulation

In the United States it is common that local governments subsidize windstorm insurance. This obviously has an effect on the market. In addition to this, an even greater effect is imposed by governmental insurance regulation. Insurance is the largest state regulated industry; each of the 50 states regulates the insurance industry individually with an agency called “insurance commission” (Meier, K.J 1991: 700, 701). Insurance commissioners are granted several mechanisms to influence the industry^[33]. While insurance represents the most profitable industry, it is also the most complicated to handle. “A large number of state insurance commissions quite frankly lack the regulatory resources to be an effective regulator of the insurance industry… (Meier1991: 703).” The federal government has often tried to gain more influence on state regulation but has seldom succeeded. In 1945 a bill was passed delegating insurance regulation to the states, “except in instances where federal law specifically supersedes state law” (Klein 1998: 27). This is the case for floods in the United States. Insurance for natural disasters composes a large part of the industry. Regulation of insurers of natural disasters differs among states, some impose binding constraints on premium rates, while other states leave the market to determine rates (Kunreuther et al. 2008a: 36). “For example, regulators and legislators in states with a high level of catastrophe risk, such as Florida and California, have faced strong political pressure to enforce extraordinary measures to constrain the price of catastrophe insurance and ensure its availability (Klein 1998: 19).”

Regulating politicians aim to ensure that insurance is available, expecting political votes in return. They also want to keep insurance in high risk areas affordable. This is enforced by rate caps on premiums and restrictions on policy cancellations. These regulations cause problems. Firstly, insurers struggle to reflect risk in their premiums; therefore the populations of high risk areas are artificially high. Secondly, paradoxically, insurers prevented from reflecting risk in premiums in a state tend to leave the market in that state, consequently lessening the availability of coverage (Kleindorfer et al. 2007: 15; Kunreuther et al. 2008a: 9)^[34].

The shortsightedness of politicians is the explanation behind these binding regulations on insurers. Politicians often only view the risk of disaster occurring in their present term of office, thereby ignoring the long term effect of binding regulation as insurers cannot amass sufficient reserves. Politicians gamble with risk and expect the earliest strike of disaster to be after their term of office. In doing so, they postpone responsibility and leave following politicians to deal with risks. This is referred to as the “Not in my Term of Office” phenomenon (Kunreuther et al. 2006b: 2). Some states seem aware that the resources private insurers dispose of are not sufficient. Politicians of coastal areas under risk of hurricanes exacerbate (i.e. they restrict the resources even more) this by allocating wind pools that provide subsidized coverage for storms. Seven of the most endangered states run a dedicated residual market mechanism for storm insurance^[35]. Participation in these residual markets is mandatory for insurance companies writing policies in these states. Insurers cover for deficits in the wind pool after a hurricane strikes which is seen as a kind of excise tax (Sutter 2009: 14). In addition to wind pools the state of Florida created a “Florida Hurricane Catastrophe Fund” (FHCF) in 1993. This functions as a reinsurance resource enabling insurers to pass on risk (Grace et al. 2001: 38). Funds like these create a moral hazard for insurers and consumers as incentives to reduce financial risk are diminished (Klein 1998: 23). For the case of insolvent insurers each state operates a guaranty fund ensuring payment for claims. Guaranty funds are operated independently. Scholars generally agree that the insurance industry is too complex to regulate efficiently. While some regulations may be harmful others may support the market. To generalize, one can divide insurance regulation into two major goals:

I. Market regulation
II. Solvency regulation

In most states the majority of premiums are written by non-domestic insurers, therefore regulators rather concentrate on the solvency of domestic insurers (Klein 1998: 34). By removing regulations that cause market difficulties regulators can also stimulate the market by two means. Firstly by ensuring insurers meet their contractual obligations, thus protecting uninformed consumers (Klein 1998: 23). Secondly, they can provide easier access to information on price transaction costs for buyers and insurers (Klein 1998: 24).

2.4 Interaction between Institutions

The system for dealing with windstorms in the US has clearly defined roles. Private insurers provide coverage. Government imposes regulation on insurers. Regulations aim to maintain availability and affordability. Regulators also enforce building codes to support the insurability of coastal areas. FEMA aids uninsured individuals and provides disaster aid for uncovered domains. The ideal functioning of this system is as follows: Private insurers provide coverage for the financially able. FEMA compensates for individuals living in non-prosperous circumstances. By retaining availability and minimizing premiums regulators hold a state of balance between two populations. On the one side is the population able to afford hazard insurance and on the other side are those relying on federal compensation. Under optimal circumstances the institutions within the system cover all sides. After this theoretical view of the system a major setback for the system dealing with windstorms in the US is examined below.

2.5 Hurricane Katrina

Hurricane Katrina struck the Gulf of North America on August 29th 2005. Most severely affected states were Louisiana, Mississippi, Alabama and South Florida. Katrina was the second most devastating hurricane ever to strike the US. Fatalities amounted to 1,323 (Swartz et al. 2006: 3). Over 1.5 million people had to be evacuated from the affected area (Kerjan 2008b: 822). Katrina destroyed 147.000 structures and 100 offshore oil and gas-platforms. Total damage was estimated to be between $80 and $130 billion (Swartz et al. 2006: 6). Other estimates assess the economic damage at between $150 and $200 billion (Kerjan 2008b: 822). Insured losses amounted to $40 billion dollars in 2005 (cf. table 1). The damage caused by Hurricane Katrina resulted in the largest insurance losses in US history (Swartz et al. 2006: 6). The amount of disaster aid spent on Katrina is hard to assess. However, by the end of 2005 about $67. 9 billion was made available^[36] (Swartz et al. 2006: 6). The 2005 season had seen three major hurricanes striking the Gulf of Mexico within 6 weeks. Katrina, Rita and Wilma together provoked disaster relief of appoximately $140 billion. Following Katrina discussion on the sustainability of the system was widespread, and it posed the question of how do deal with future disasters. Problems arising from this system are discussed in part two of this thesis.

illustration not visible in this excerpt

Table 1: The costliest US Hurricanes of the United States – adjusted to insured losses (2005 Dollars)

3. Droughts in the United States

Over the previous decades drought has been a cause of major financial losses within the United States. Drought as a natural hazard differs from floods and storms in several ways. Apart from the fact that drought occurs gradually over a long time rather than by a sudden impact, drought is measured and perceived quite differently to other natural hazards. In dealing with this topic the main lines of opinion on drought are set out and the common methods of measuring drought are described. This suggests that farmers are the people most heavily affected by dry periods (3.1). Primary institutions managing drought are governmental organisations. The reason for this is that the American government developed a state insurance scheme in 1938. Roles of these organisations within the drought management context are discussed (3.2.)

[...]

---

^[1] Insured losses in the 1970s amounted to averagely $3billion, whereas insured losses in the first decade of the 21st century average at $34 billion (indexed to USD 2007) (Swiss Re 2008). In turn, total losses averaged at $16 billion in the period from 1970–1979 compared to an average of 77 billion in the period from 2000-2008 (Kunreuther 2009: 4). Total losses have quintupled, while in the same period insured losses have seen a twelve folded increase. This fact asks for explanation. The most obvious solution is that insurers have failed to assess risks. However, the more common view is that in the recent decades weather related disasters have tripled and that society has continuously settled in hazard prone areas.

^[2] Within a normal environment an insurer has an informational advantage on risks, therefore able to distinguish between good and bad risks. This allows the insurer to include both risks in a similar extent, if this informational advantage is twisted then it is a case of adverse selection “In an extreme case, the poor risks will be the only purchasers of coverage, and the insurer will lose money on each policy sold. This situation, referred to as adverse selection, occurs when the insurer cannot distinguish between the probabilities of a loss for good- and poor-risk categories. (Freeman 2007: 172)”

^[3] Organisations commonly apply problem shifting as a measure, where solutions are difficult to find (Kieser et al. 2006: 453).

^[4] This increases the threat of suffering from an adverse selection effect.

^[5] These disturbances are merely examples, all effects and constraints withholding a system from functioning in intended fashion are classified as “disturbances”. As the concept of risk clarifies, virtually everything may be viewed in respect to the extent of risk involved; as in this case everything may be seen as a potential disturbance. Or as Richard Ericson et al. 2003 points out, “No one can escape being subject to risk discourse and risk management (Ericson et al. 2003: 68)”.

^[6] The difference between institutional arrangements singly managing one hazard and those dealing with three is notable. For example - it is thinkable that institutions may pursue a similar policy for all natural hazards; insurers may redline a populated area regardless of hazard type. This suggests that comparison of various countries and hazards types is necessary to distinguish between and general and specific policies. „Redlining is the term for unfair discrimination, not based on real differences in risk or cost, against a particular geographic area (Ericson et al. 2003: 64).”

^[7] Federal Emergency Management Agency (FEMA) will be more closely viewed in the chapter on Windstorms in the US.

^[8] http://www.fdic.gov/regulations/laws/rules/6500-3600.html Last accessed: 26.12.2009

^[9] 5/13/2008--Passed Senate amended. Title I: Flood Insurance Reform and Modernization - Flood Insurance Reform and Modernization Act of 2008 - (Sec. 104) Amends the National Flood Insurance Act of 1968 (Act) to extend the national flood insurance program (Program) though FY2013. (Sec. 105) http://www.govtrack.us/congress/billtext.xpd?bill=h110-3121 last accessed 26.12.2009

^[10] http://www.govtrack.us/congress/billtext.xpd?bill=h110-3121 last accessed 26.12.2009

^[11] http://www.fema.gov Last accessed 26.12.2009

^[12] http://bsa.nfipstat.com/reports/1011.htm Last accessed 26.12.2009

^[13] http://www.fema.gov/business/nfip/statistics/cy2007lsdoll.shtm Last accessed: 04.09.09

^[14] http://www.fema.gov/business/nfip/statistics/sign1000.shtm Last accessed: 06.06.09

^[15] http://www.fema.gov/femaNews/disasterSearch.do?pageInfo.pageStart=1176 Last accessed: 26.12.2009

^[16] http://www.sba.gov/services/disasterassistance/index.html Last accessed: 06.06.09

^[17] http://www.sba.gov Last accessed: 13.07.09

^[18] http://www.sba.gov/services/disasterassistance/basics/Mission/index.html Last accessed: 13.07.09

^[19] http://www.fema.gov/femaNews/disasterSearch.do?pageInfo.pageStart=1176 Last accessed: 7.1.09

^[20] http://www.fema.gov/library/viewRecord.do?id=1935 Last accessed: 7.1.10

^[21] This law stated that congress would continue their means of assistance. Main goals were listed as: broadening existing relief programs, achieving greater coordination between disaster programs, encouraging protection programs to supplement disaster aid, encouraging hazard mitigation http://www.fema.gov/pdf/about/stafford_act.pdf Last accessed: 7.1.10

^[22] http://www.fema.gov/library/viewRecord.do?id=1935 Last accessed: 7.1.10

^[23] The decision, whether or not to request disaster aid is left entirely to a state’s Governor. Therefore leaving the governor to assess whether the disaster will go beyond what the state can handle autonomously.

^[24] The standard procedure is for the President to consult with FEMA whether or not declaring disaster is appropriate.

^[25] http://www.fema.gov/femaNews/disasterSearch.do?pageInfo.pageStart=1176 7.1.09

^[26] The Write Your Own (WYO) Program began in 1983 and is a cooperative undertaking of the insurance industry and FEMA. The WYO Program allows participating property and casualty insurance companies to write and service the Standard Flood Insurance Policy in their own names. The companies receive an expense allowance for policies written and claims processed while the Federal Government retains responsibility for underwriting losses. The WYO Program operates as part of the NFIP, and is subject to its rules and regulations http://www.fema.gov/business/nfip/wyowhat.shtm Last accessed: 01.09.09

^[27] http://www.fema.gov last accessed: 12.08.09

^[28] “A major hurricane has a rating of 3, 4 or 5 on the Safir-Simpson scale of hurricane intensity. The scale rates hurricanes from 1 to 5. The winds of a category 3 hurricane are 111 mph or greater (Sutter 2009: 9).”

^[29] http://www.fema.gov/femaNews/disasterSearch.do?pageInfo.pageStart=276 Last accessed: 11.6.09

^[30] The PDA is a combined effort from state, local and FEMA officials (Faure et al. 2003: 316).

^[31] The purpose of hazard mitigation assistance is to reduce or eliminate an area’s vulnerability to similar disasters in the future (Faure, M et al. 2003: 317).

^[32] “ERT-As are located in the ten FEMA Regions and can be deployed in the early phases of an incident to work directly with the states to assess the disaster impact, gain situational awareness, help coordinate the disaster response, and respond to specific state requests for assistance. The ERT-As are made up of approximately 25 individuals from the FEMA Regions, who also have day-to-day responsibilities beyond their team assignments, and representatives from the ESF Departments and agencies (http://www.fema.gov/media/archives/2007/061207.shtm Last accessed: 03.09.09).”

^[33] “State insurance commissioners have the principal authority to regulate the business of insurance within their jurisdictions. This authority encompasses: licensing insurers and producers; oversight of insurers’ financial structure, investments, transactions and operations; approval of insurance products and prices; and policing insurer and producer trade practices (Klein 1998: 27).”

^[34] The analysis of the supply of coverage indicates that if regulators suppress rates too much, there is likely to be a severe decrease in the availability in coverage (Kunreuther et al. 2008a: 9).

^[35] To make basic coverage more readily available to everyone who wants or needs insurance, special insurance plans, known as residual, shared or involuntary markets, have been set up by state regulators working with the insurance industry (http://www.iii.org/media/hottopics/insurance/residual/?table_sort_746398=-3&printerfriendly=yes Last accessed: 16.07.09).

^[36] An additional $1.1 billion was made available for ongoing students located in affected areas (Swartz et al. 2006: 6).