Flood Risk Exposure in Austria – Options for Bearing Risk Efficiently

Url, Thomas; Sinabell, Franz Article Flood Risk Exposure in Austria – Options for Bearing Risk Efficiently Schmollers Jahrbuch – Journal of Applied Social Science Studies. Zeitschrift für Wirtschafts- und Sozialwissenschaften Provided in Cooperation with: Duncker & Humblot, Berlin Suggested Citation: Url, Thomas; Sinabell, Franz (2008) : Flood Risk Exposure in Austria – Options for Bearing Risk Efficiently, Schmollers Jahrbuch – Journal of Applied Social Science Studies. Zeitschrift für Wirtschafts- und Sozialwissenschaften, ISSN 1865-5742, Duncker & Humblot, Berlin, Vol. 128, Iss. 4, pp. 593-614, https://doi.org/10.3790/schm.128.4.593 This Version is available at: https://hdl.handle.net/10419/292243 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Flood Risk Exposure in Austria – Options for Bearing Risk Efficiently By Thomas Url and Franz Sinabell* Abstract Due to its topography, Austria is exposed to many natural hazards. Inadequate spatial planning aggravates natural exposure to risks, 12% of all buildings are potentially exposed to flooding almost 9% are considered to be at an extreme risk. Reinforcing precautionary measures is a prerequisite for an efficient risk management system that will be ready to meet future challenges. In an efficient risk management system the Austrian government’s relief payments after catastrophic events should be substituted by a broad (mostly compulsory) insurance against natural hazards. The strong involvement of government in the provisions of precautionary measures against natural disasters and its role during emergencies should be better co-ordinated with private measures. JEL Classifications: Q54, G28, G22 1. Introduction Among the natural hazards that occur most frequently in Austria are floods, avalanches, storms, snow pressure and hail. In the past few years major flooding events caused significant damages (in 2002 about A2.9 billion and in 2005 about A0.6 billion). Earthquakes also present a great danger (above all in the south of Carinthia), but they rarely happen. Volcanic eruptions, storm flooding and tsunamis are major international threats that, however, do not directly affect Austria due to its geographical location. In industrialised countries such events mainly cause property damage, but particularly in developing countries, the death toll runs high every year. Natural disasters happen less frequently than catastrophic events caused by people (such as terrorist attacks, chemical accidents) but the damage (number of victims, property losses) is higher (Swiss Re, 2006). Schmollers Jahrbuch 128 (2008), 593–614 Duncker & Humblot, Berlin Schmollers Jahrbuch 128 (2008) 4 * This paper is based on results of a research project funded by the Austrian Insurance Association VVO. Efficient research assistance by Urula Glauninger and Dietmar Weinberger and valuable comments to a previous version of this paper by Wilfried Puwein, Peter Mayerhofer, participants of the ‘Workshop on Financial Risk of Natural Hazards- Markets and the Role of the State’ on July 5, 2007 in Innsbruck, and an anonymous referee are gratefully acknowledged. All remaining errors are ours. OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 594 Thomas Url and Franz Sinabell On a global scale, the trend in both the frequency of the occurrence of natural hazards and the damage of the events is rising. Changes in the world’s climate nurse the fears that society will need to brace itself for more frequent and more serious climatic disasters. Human behaviour influences these natural processes to the worse (IPPC, 2007) and is at the same time resulting in an increasing vulnerability to natural hazards. Two developments are primarily responsible for this in developed countries: economic growth, which results in ever higher goods and property values, and the expansion of settlements and infrastructure area into at risk regions. In the less-developed countries these processes are eclipsed by the strong population growth. Even if we assume that natural risk exposure did not change over time, this means that the economic damage potential is rising: In the last decade of the last century the per annum real growth rate of the capital stock of the Austrian national economy was 2.6% (Schwarz, 2002). The sealed area of permanently populated space in Austria reached almost 6% in the year 2006. The increase in sealed area has remained constant for years at a daily level of land consumption of about five hectares, although the Austrian strategy for sustainability targets a level of one hectare per day (UBA, 2007). The question is how private and public players can take these developments into account in their plans and how they should adapt their behaviour correspondingly. Part of the efficient handling of natural hazards is keeping the level of damage low and, despite a certain risk exposure, to undertake as many economically profitable activities as possible. In this paper we will demonstrate how this goal can be achieved through a coordinated interweaving of public and private solutions. Starting point for our considerations is the current risk management system for dealing with floods in Austria. We will focus in detail on flooding because this natural hazard has caused considerable damage over the last few years. In the following sections we review the literature to identify the preconditions of an efficient system for the management of natural hazards from an economic point of view. Then we will make a proposal to optimally meet the following requirements during the three phases before, during, and after a natural hazard occurs: in the first phase, farsighted damage prevention steps must be taken before a natural phenomenon strikes; in the second phase, as the event goes on, mitigation measures must be realised quickly; in the third phase after the natural disaster, repairs and reconstruction has to be financed and the persons affected must be compensated at terms that are known in advance. An efficient risk management system helps ensure that steps are taken in all three phases to keep the total extent of damages as low as possible. An important instrument for doing so is the risk transfer system which distributes the financial consequences of uncertain events within a collective. Although this is only fully felt after the damage has occurred, the concrete characteristics have a great influence on the efforts potential victims make to prevent and reduce damages. Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 Flood Risk Exposure in Austria 595 2. Investments in Flood Protection and Flood Exposure in Austria Numerous means of preventing natural hazards are actually pure public goods (cf. Mueller, 2003, 18 ff.). Such goods are characterised by the fact that the exclusion of other households from consumption is only possible at a high cost or not at all. Theoretically, the value of public goods is derived from the willingness to pay of those households that benefit from this good. Another characteristic of public good provision is that private households behave strategically. They indicate a low willingness to pay for a public good and can thus keep their financial contribution (e.g. for a dam) low. Therefore, the market does not provide public goods at all, or only insufficiently. This free riding behaviour on the side of private households can only be overcome through a public choice mechanism about the volume of the public good in question. The Austrian legislature has solved this challenge in such a way that only a small part of the costs of preventive protective constructions is borne by the beneficiaries. Federal and provincial governments pay the largest share (up to 80 % of total costs) of preventive measures. More than 200 Mio Euros have been spent annually on dams, plans and information systems that are important for assessing the hazards caused by natural phenomena during the last years. In addition, mitigating facilities such as fire brigades are co-financed by the public so they are able to undertake disaster rescue missions more effectively. There are detailed hazard zone maps for almost all the areas that are exposed to torrents and avalanches (die.Wildbach und Lawinenverbauung, no date). They show the different degrees of risk exposure and the processes that cause them. The corresponding maps are filed in the municipalities and form an important base for decisions about which areas should be excluded from certain uses. Analogue hazard zone maps are also being worked out for rivers but they do not cover all the territory yet. The municipalities play an important role in damage prevention. By dedicating zones as being suitable for development they determine where buildings and infrastructure can be built. Hazard zone maps are information systems that are relatively difficult to put together and working them out requires a great deal of time. One result of this is that the proper plans were not available in the past and local zoning did not give enough consideration to flood risk exposure. In addition, investigations following the floods of 2002 (cf. Habersack et al., 2004) discovered that existing maps had not been sufficiently considered in the local zoning regulation, meaning that the number of properties at risk of flooding in Austria is very high. This was identified as one reason for the high level of flood damages in 2002 (cf. Sinabell / Trimmel, 2004). To find out more about the extent of flood risk exposure, HORA (Austrian flood risk zones, “HOchwasserRisikoflächen Austria”) was published in the Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 596 Thomas Url and Franz Sinabell summer of 2006. It is funded by the Federal Ministry of Agriculture, Forestry, Environment and Water Management and the Austrian Insurance Association (VVO). With this information system it is now possible to better show the degree of flood risk exposure of individual objects throughout the entire territory. At the moment, the system is still lacking the precision of hazard zone maps. The reduction of flood risk due to dams and other protective measures is not yet accounted for. For many questions, however, this disadvantage is offset by its widespread availability and its accessibility over the internet with minimal efforts. The extent of flood risk exposure in Austria is shown in Figure 1, disregarding protective constructions like dams. It shows the potential risk exposure on a municipal level if protective steps that have already been taken should fail, for example if a dam breaks. The different colours of the municipalities depict how many properties (buildings with valid addresses) in the respective municipality are within the zones of 200-year flooding events. The darker the colour, the higher the portion of properties that is at risk. In Table 2 the corresponding figures are summarised on the level of federal provinces (Länder). Additional information like settlement area and the percentage of sealed surfaces (streets, buildings, etc.) at the level of provinces shows that the number of properties in risk zones is very high in mountainous provinces where settlement concentrates along rivers in the valleys. S: Land-, forst- und wasserwirtschaftliches Rechenzentrum GmbH (2006), own results. Figure 1: Exposure of public, commercial and private properties to flood risks in Austria in 2005 Altogether, slightly more than 242,000 properties are regarded as potentially at risk (they are within the zone of 200-year floods, should protective constructions fail). This corresponds to about 12% of the total. The majority of Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 Flood Risk Exposure in Austria 597 these (8 %) are in the zone of 30-year events (zone 1 in Table 1), meaning they are at a potentially high risk. In Austria, despite the build up of HORA, information is not being gathered systematically about the economic consequences of natural disasters. Although the number of properties at risk is now known well enough, we still do not know what the economic value of these properties is. The only source published regularly is the report of the disaster fund which finances public relief payments after catastrophes and precautionary measures. However, this report only lists the sum of the federal grants that have been awarded, without distinguishing between event categories (e.g. floods, avalanches) or breaking down the total damages. In general, knowledge about the economic effects of natural disasters is therefore very limited. Exceptions are the floods of 2002 and 2005 that are well documented (Habersack et al., 2004; Rudolf-Miklau, 2006, Sinabell / Url, 2006). Studies on these events allow conclusions as to what portion of private damages is covered by state aid, insurance and private donations. Across all the claimants, an average of about 80 percent of their private damages has been covered; i.e. the “insured’s share of the risk” amounted to about one fifth of economic losses. Around this mean there is substantial variation in the coverage rates. For some households the share of the damage they had to carry was existence-threatening, on the other hand, depending on local governments, it was also possible to have the damage fully compensated. Table 1 Investments in flood prevention measures in Austria real annual expenditures (average 2001–2005) at prices 2005 Länder total Mio. A per household A per capita A per Mio A gross value added per object in a hazard zone 1 A Burgenland 20 191 71 4,206 2,387 Kärnten 23 104 41 1,904 1,465 Niederösterreich 28 47 18 884 383 Oberösterreich 28 52 20 841 768 Salzburg 26 128 51 1,784 1,329 Steiermark 28 62 24 1,102 1,056 Tirol 30 118 45 1,749 1,376 Vorarlberg 17 128 49 1,840 1,096 Wien 19 25 12 342 779 Austria 219 67 27 1,072 906 Sources: Estimates based on Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft (2006A); Landesrechnungsabschlüsse (Budget-Ansatz 1 /63). 1 Hazard zones according to HORA. Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 598 Thomas Url and Franz Sinabell Since different institutions are entrusted with the processes, and flooding can be caused both by natural streams that break their banks as well as by torrents and other waterways, it is not always possible to clearly allocate the steps that need to be taken for preventive flood protection. One example is that measures to improve rivers for shipping traffic are carried out simultaneously with flood protection improvements. The numbers presented are therefore confined to public investments that can be assigned unambiguously. In Table 1, the expenses cited for flood protection measures are also compared to the number of properties that HORA identifies as being potentially at risk. The table shows that funds for preventive measures are not distributed evenly across the country. The largest expenditures per household or per capita are in the region with the lowest number of properties in flood risk zones (Burgenland) whereas investments in preventive measures are relatively low in the province with the most properties in risk zones (Niederösterreich). Table 2 Number of properties in hazard zones and sealed up area in the Austrian La¨nder 2005 properties settlement area properties in risk zones 1 total in hazard zones total 1,000 km 2 sealed up 2 share in % zone 1 to 3 share in % zone 1 (high risk) share in % number Burgenland 114,831 8,254 2.5 5.2 7.2 5.6 Kärnten 150,708 15,594 2.3 6.8 10.3 8.8 Niederösterreich 545,801 73,531 11.3 5.0 13.5 9.7 Oberösterreich 354,861 35,755 6.6 5.0 10.1 7.6 Salzburg 114,330 19,732 1.4 7.0 17.3 13.9 Steiermark 319,083 26,785 5.0 6.6 8.4 6.2 Tirol 153,196 22,044 1.5 7.9 14.4 11.4 Vorarlberg 88,181 15,527 0.6 8.0 17.6 15.5 Wien 174,407 24,829 0.3 28.0 14.2 6.2 Austria 2,015,398 242,051 31.5 5.9 12.0 8.8 Source: Statistik Austria, Gebäudezählung 2001, Land-, forst- und wasserwirtschaftliches Rechenzentrum GmbH (2006); Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft (2006A). According to “Hochwasserrisikozonierung Austria – HORA” properties are potentially at risk (“potentiell gefährdet”) ignoring protection due to dams and other constructions (Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, 2006A, detailed definitions are available at http: //www.hochwasserrisiko.at. 1 Number of properties in zone 1 (high risk: expected T=30-year flood level). Lower risks are in zone 2 (T=100-year flood level) and zone 3 (T=200-year flood level). 2 Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft (2006B). Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 Flood Risk Exposure in Austria 599 Sinabell / Trimmel (2004) examined the factors on which decisions were based on when judging specific protection measures. They found that the criteria used to judge the cost-benefits of river engineering projects in the year 2004 did not reflect the standards of cost-benefit analysis. The estimate of the costs of projects was relatively reliable. However, the assessment of the benefits of river engineering projects was flawed because estimates were often based on standardised rates without recognising actual conditions. In addition, project assessments did not take external effects on downstream riparian into account. 3. Risk Management and the Risk Transfer in Austria There are some obvious imperfections in Austria’s current risk management and risk transfer systems (cf. Sinabell, 2004; Hyll / Vetters / Prettenthaler, 2004; Prettenthaler / Hyll / Türk / Vetters, 2004). Particularly the coordination between risk management and the risk transfer system appears inefficient. In addition, the risk transfer system exhibits the following deficits: The majority of private households is insufficiently insured against damages caused by natural hazards (including floods). In some Austrian provinces the indemnities paid by private insurances reduce the amount of public disaster relief granted. This lessens people’s incentive to take out private insurance policies. Many people who would like to take out insurance but live in risky zones (such as areas that were flooded within the past five years) do not find an insurance company that is willing to cover the risk completely. If insurance policies do offer coverage it is usually restricted – either to a percentage of the sum insured (e.g. 50 %) or a flat payout (between 4,000 and 7,000 Euros). Insurance companies set an upper limit to their total indemnity by including clauses against accumulated risks (e.g. 15 million Euros per event). This means the actual claims payments depend on the total number of people affected by a flood and are therefore not known in advance. Summing up, low cost damages occurring more frequently are fully covered by standard household policies while large damages occurring less frequently are incompletely covered. These observations indicate a market failure. On the demand side households in risk zones only look for insurance when they are – subjectively – worried enough. One reason for the low demand could be the subjective underestimation of the actual flood risk. HORA should have removed this problem, because this information system provides a risk assessment concerning floods and earthquakes in the form of easily comprehensible T-year flood levels. Yet a lack of information search or limited interest still may lead to a huge gap between perceived and objective risk. Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 600 Thomas Url and Franz Sinabell On the supply side, the insurance industry is mainly selling contracts which leave the policyholders extremely underinsured (flat sum). An important reason for this lies in the nature of risks from flood. Floods affect a large number of people simultaneously. Large events thus can overstrain the liquidity of individual insurance companies because the risk can only partially be diversified within the insurance pool and reinsurance is costly (Froot, 2001). In Austria the federal government makes up for this market failure by providing disaster relief to victims. Over the last few years roughly 80 million Euros in state aid have been paid out annually to claimants (Sinabell / Url, 2006). Financial assistance granted from the disaster fund and the budgets of the provinces are funded by tax money. The deficiencies above show that the Austrian government’s efforts to remedy low insurance coverage by taking over the risk transfer system have only been partially effective. Some 40 percent of private losses have been covered by public disaster relief after recent floods but there is no legal entitlement to claim compensation, and the share of damage after accounting for private insurance indemnities and relief payments is still very high in individual cases. Another weakness is the missing link between the risk transfer system on the one hand, and the risk exposure of households on the other. This means that an important criterion of an efficient risk transfer system, i.e. a risk adequate financing of indemnities is not fulfilled. This problem is reinforced by the fact that some federal provinces offset private insurance payouts against public relief payments. As a consequence, households do not spend enough effort to avoid damages and take out too low an insurance coverage. Before presenting a proposal how to modify the Austrian risk transfer system for natural hazards we will briefly discuss requirements for an efficient risk transfer system in the case of large scale events. 4. Requirements Placed on an Efficient Risk Management or Risk Transfer System Risk management of natural hazards sums up all the measures for information and prevention or for damage reduction before and during a natural disaster. The risk transfer system, on the other hand, only describes the transfer of individual risks to risk collectives such as insurance companies or public relief funds. This can be a transfer system financed by taxes, as in the case of the disaster fund or, as in the case of storm insurance, an insurance system financed by premium payments. The Austrian hail insurance can be described as a mixed system because private premium payments are subsidised by public transfers. Risk management and risk transfer are not independent of each other. The instruments from the two areas can either strengthen or weaken each other. Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 Flood Risk Exposure in Austria 607 On the basis of this experience, we propose a withdrawal of the government from the risk transfer system, which leaves open two alternative options: (1) The potential claimants carry the costs of the damages themselves by accepting self-insurance or self-protection in the sense of Ehrlich / Becker (1972), or alternatively (2) a system with compulsory insurance with private insurers taking over the risk transfer at risk adequate premiums. Option (1) suffers from a lack of credibility because the government has no instrument to commit itself credibly not to pay out disaster relief after an incident, cf. Coate (1995). A purely market based solution will involve low levels of self-insurance and selfprotection by private households because of the possibility of charitable transfers from altruistic rich households in the bad state. In this case, there is no reason to expect the level of transfers to be optimal for victims. Additionally, there is a free rider problem in fixing the level of charity payments by rich households, and there is no way to implement cost-effective protection in the presence of self-insurance opportunities. Moreover, self-insurance does not enable the household to benefit from the law of large numbers offered by an insurance pool and opportunities to spread risks are foregone. A system of compulsory insurance with risk adequate premiums, on the other hand, provides the government with a commitment device (Kunreuther / Pauly, 2006), generates the biggest possible insurance pool, and induces complementary self-protective action, although self-insurance will be substituted by compulsory market insurance. Insurance systems based on risk adequate premiums have the advantage that complementary ex ante damage prevention and damage-mitigating activities are triggered through risk adequate premiums. The practical instruments for this are differentiated premium levels, discounts, deductibles, and on-site inspections. Risk-adequate premiums increase the risk awareness of private households because properties that are in more danger or less protected also face higher insurance premium. This information already influences the choice of the location where properties may be built and should also lead to a situation where the benefits of all ex ante protective measures (anchoring oil tanks, for example) are balanced against their costs. Risk-adequate premiums reward efficient protective measures, depending on how much it costs the insurer to monitor them, either with discounts (ex ante) or with deductibles (ex post). In so doing, they contribute to the efficient reduction of the risk of natural hazards (Shavell, 1979; Winter, 2000). Alternative instruments are experienced based bonus malus systems, partial insurance exclusion or coinsurance clauses. Insurance policies with a coinsurance clause only cover a portion of the damage, starting at a damage amount that is agreed upon in advance. None of these instruments requires monitoring the policyholder, and they automatically sanction behaviour that does not minimise damage by increasing premiums or restricting benefits. Enforcing private households to sign insurance policies does not solve the high costs of reinsurance in a reserve based insurance system. On the contrary, Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 608 Thomas Url and Franz Sinabell primary insurers are likely to pass through high reinsurance premiums into the price. For this reason we suggest establishment of a natural hazard insurer based on the mutuality principle, maybe combined with some degree of reinsurance. Underwriting non-diversifiable risk can be reduced with reinsurance, for which a risk premium has to be paid. This organisational form is superior to other forms because it can carry both diversified and non-diversified risk to a large degree. The mutuality principle makes upper limits for coverage superfluous. Limits to the total indemnity violate an essential element of an efficient risk transfer system because complete coverage is not offered under foreseeable conditions. If the government acts as the insurer of last resort it will effectively subsidise the risk premium and thus violate the principle of riskadequate premiums. 5.1 Transition to the New Risk Transfer System Several factors impede an unmodified implementation of our proposal in Austria. First of all, there are legal hurdles against a system of compulsory insurance. Partial coverage against flood insurance, as already offered by some firms, is an indication that the market does not fail completely and EU-law limits the implementation of compulsory systems if market institutions already exist. To establish a system similar to that in Spain or Switzerland bears considerable legal risks. A system of compulsory insurance which had to be abandoned after being challenged at courts would do no good. A way out of legal challenges could be achieved by systems similar to those in the UK, in Italy or Belgium where insurance of natural disasters is bundled with fire insurance. Buying coverage against natural hazards would then not be compulsory in a legal sense but quasi-compulsory for most owners of properties. Besides legal objections, political resistance may be vigorous. The change would result in a shift of the burden from public tax-financed transfers to those households exposed to moderate or high risk properties. Several elements of an insurance based solution could help reduce this: Insurance premiums should only reflect risks, i.e. there should be no redistribution between risk classes. By pooling different uncorrelated natural perils the number of households with insurance would rise and premiums could be offered less expensively. Compulsory insurance would probably be more acceptable for many of the people if the tax burden were reduced at the same time compensating for abandoned funding of public catastrophe relief. Nevertheless, the group of persons that is likely to be affected most by compulsory insurance (the owners of approximately 200,000 high-risk objects) is relatively small compared to the group of persons that would benefit from such a solution (several million tax payers). In many cases smaller interest groups are more effective in manipulating the policy making process in their interest (Olson, 1982). Therefore it is possible that the public savings of approximately 80 Mio. Euros per annum Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 Flood Risk Exposure in Austria 609 will not suffice to compensate the majority of low-risk households for their costs of supporting an insurance based solution. On the supply side, insurers will have a vital interest to transfer part of the risk to the state as the insurer of last resort. Schwarze / Wagner (2007) mention this as one of the decisive reasons for the failure of establishing a compulsory insurance based system in Germany. The mutuality based system would be able to cope with a lower degree of reinsurance as compared to a reserve based system and thus faces less need for a state layer covering large scale disasters. Nevertheless, subsequent profit or loss sharing within the insurance pool might be challenged in the case of a large scale disaster and an additional state layer for extreme events may be considered, e.g. as a returnable public credit to the mutual (Jaffee / Russell, 2006). Such a construction should attenuate resistance by the government against a role as an insurer of last resort. Homeowners in high risk zones have a high interest to lobby for premium subsidies. Some people might find it ’fair’ that those in zones without any risk contribute to a collective burden sharing system because of equity concerns. It is therefore not unlikely that in one way or another, premiums will be subsidised (either by the tax payers or by low-risk agents). A negative consequence of granting premium subsidies is reduced incentives to avoid damages. Subsidies for premiums, however, seem to be the most suitable instrument for mitigating the burden of the compulsory insurance on disadvantaged groups while avoiding the inefficiencies of charity payments by private households (Coate, 1995). Social equity concerns are likely to emerge as poor households tend to settle in hazard prone zones because of lower land prices. If premium subsidies have to be granted due to equity concerns, regional administrative bodies (primarily the local communities) should finance them because they were and are responsible for designating zones for developments. However, if there is a premium subsidy it should only be temporarily granted and limited to existing properties. New developments would carry the price signal of high insurance premiums with them and thus ensure risk adequate behaviour. 5.2 An Appraisal of the Political Feasibility of the Proposed Risk Transfer System The bundling of natural hazards with fire insurances has been suggested by several authors as a means to improve the risk transfers system in Austria and Germany. This would be a quasi-obligatory insurance while we propose mandatory insurance of all non-commercial properties in hazard zones. From an administrative perspective it is advantageous to add coverage to an existing contract which may increase acceptance by private households. But if fire insurance is not comprehensive we again end up with uncovered damages and consequently incentives for the public to provide disaster relief. These incentives grow with the value of properties in natural hazard zones lacking fire Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 610 Thomas Url and Franz Sinabell insurance. Therefore, an evaluation of this exposure is necessary prior to the decision over the implementation of a mandatory or a quasi-obligatory system. Political resistance against a mandatory insurance seems to be lower if commercial properties are not included because the heterogeneity of commercial risks could make the calculation of premiums less transparent. Authors like Schwarze / Wagner (2007) and Prettenthaler / Vetters (2005) propose an alternative organisation of the risk transfer system: primary insurers underwrite contracts with clients, provide claims management services, and pool the risk of natural hazards in a first layer. In both proposals, the state is the insurer of last resort which allows saving the risk premium for reinsurance. Since the mutuality principle also minimizes the need for costly reinsurance private households should be indifferent with respect to both approaches. Reinsurance by the state transfers wealth from tax payers to households in risk zones and dampens efforts to prevent damages. Furthermore, it creates unforeseen pressure on public budgets. Prettenthaler / Vetters (2005) plead for a “risk differentiated premium” but oppose a risk adequate premium because of the high costs in high risk zones. This system implies subsidies flowing from low risk towards high risk households. The acceptance of such a system is likely to be low and a preference for redistribution is better implemented by temporary targeted subsidies for low income households. 6. Discussion and Conclusions The European Floods Directive will create an EU framework for flood risk management (CEC, 2006). It builds on and is closely coordinated and synchronised with the 2000 Water Framework Directive, the cornerstone of EU water protection policy. The Floods Directive adopted by Council and Parliament will require that Member States take a long-term planning approach to reducing flood risks in three stages (CEC, 2007): Member States will by 2011 undertake a preliminary flood risk assessment of their river basins and associated coastal zones. Where real risks of flood damage exist, they must by 2013 develop flood hazard maps and flood risk maps. By 2015 flood risk management plans must be drawn up for these zones. These plans are to include measures to reduce the probability of flooding and its potential consequences. They will address all phases of the flood risk management cycle but focus particularly on prevention, protection and preparedness, e.g. providing instructions to the public on what to do in the event of flooding. Results presented in this paper show that Austria spends large sums for preventive measures in regions with relatively low risks. This finding supports the view that measures are not taken in a cost-effective manner. The European Floods Directive does not seem to be an instrument with a broad enough ap- Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 Flood Risk Exposure in Austria 611 proach that forces governments of Member States to take cost-effective measures as the Water Framework Directive does. Depending on the way Member States implement the Floods Directive, national regulations may turn out to put too much weight on an engineering way of decision making confined to technical effectiveness. In this paper we stress the economics behind the three phases before, during, and after a disaster to show that a compulsory insurance based risk transfer system with risk adequate premiums fosters efficient allocation of resources over all phases of natural hazards. This favourable outcome is a consequence of complementarity between market insurance and self-protection which can be shown to hold if the premium level for an insurance contract depends on the effectiveness of self-protective measures undertaken by households. Although self-pro- tection can be expected to be implemented efficiently within our scheme, there is no mechanism to guarantee that public goods in the pre-disaster phase will be provided efficiently. Given the Austrian record of preventive spending in lowrisk areas, the next challenge after the establishment of an insurance based system of risk transfer in Austria would be to coordinate the preventive public activities, like building dams or retention basins, with signals on value at risk from the risk transfer system. The estimated value of assets under risk in hazard zones, as assessed by the insurance industry, is an essential input for efficiently spending public money. An established coordination mechanism between well informed insurers and the government would make it possible to use public funds for protective constructions in a way that is based on cost-benefit criteria. References BMLFUW (Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft) (2006A): Data on Flood Prevention Expenditures, provided by Abteilung VII / 5, as at Oct. 5 th 2006, mimeo, Vienna. BMLFUW (Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, Ed.) (2006B): Auf dem Weg zu einem nachhaltigen Österreich. Indikatoren- Bericht Juni 2006. Provided by Statistik Austria, Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, Vienna. Buchanan, J. M. (1975): The Samaritan’s Dilemma, in: E. Phelps (ed.), Altruism, Morality and Economic Theory, New York, 71 – 85. CEC (Commission of the European Communities) (2006): COM(2006)15 final as at Jan. 18 th 2006. Commission’s Communication on the Common position COM (2006)775 of 6. 12. 2006. CEC (Commission of the European Communities) (2007): Environment: European Parliament and Council reach agreement on the new Floods Directive, Press Release IP / 07 / 565, April 25 th 2007. Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 612 Thomas Url and Franz Sinabell Coate, S. (1995): Altruism, the Samaritan’s Dilemma, and Government Transfer Policy, American Economic Review 85, 46–57. die.Wildbach und Lawinenverbauung, (no date) Wildbach und Lawinenverbauung in Österreich, Informationsbroschüre, Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, Vienna. Doherty, N. A. (1997): Insurance Markets and Climate Change, Geneva Papers on Risk and Insurance 22, 223 – 237. Doherty, N. A. (2000): Innovation in Corporate Risk Management: The Case of Disaster Risk, in: G. Dionne, Handbook of Insurance, Dordrecht, 503 – 539. Doherty, N. A./ Dionne, G. (1993): Insurance with Undiversifiable Risk: Contract Structure and Organizational Form of Insurance Firms, Journal of Risk and Insurance 60, 187 – 203. Ehrlich, I. /Becker, G S. (1972): Market Insurance, Self-insurance, and Self-protection, Journal of Political Economy 80, 623–648. FEMA and FIMA (Federal Emergency Managment Agency and Federal Insurance and Mitigation Administration) (2002): National Flood Insurance Program – Program Description. Information Brochure, available at www.fema.gov / doc / library / nfipdescrip.doc (retrieved on 9 April 2008). Froot, K. A. (2001): The Market for Disaster Risk: A Clinical Examination, Journal of Financial Economics 60, 529 – 571. Habersack, H. / Bürgel, J. / Petraschek, A. (2004): Flood-Risk – Analyse der Hochwasserereignisse vom August 2002 – Synthesebericht, Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, Vienna. Hyll, W. / Vetters, N. / Prettenthaler, F. (2004): Analyse der Hochwasserereignisse vom August 2002 – Flood Risk, WP Ökonomische Aspekte TP 04, Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, Vienna. IPCC (Intergovernmental Panel on Climate Change) (2007): Climate Change 2007: The Physical Science Basis, Summary for Policy Makers, IPCC Secretariat, Genf. Available at: http: // www.ipcc.ch / SPM2feb07.pdfr (downloaded on Feb. 2 nd 2007). Jaffee, D. M. / Russell, T. (2006): Should Governments Provide Catastrophe Insurance?, Economists’ Voice, 2006, 3(5), http: // www.bepress.com / ev. Kim, B. J. / Schlesinger, H. (2005): Adverse Selection in an Insurance Market With Government-Guaranteed Subsistence Levels, Journal of Risk and Insurance 72, 61– 75. Kunreuther, H. / Pauly, M. (2006): Rules Rather Than Discretion: Lessons From Hurrican Katrina, Journal of Risk and Uncertainty 33, 101 –116. Land-, forst- und wasserwirtschaftliches Rechenzentrum GmbH (October 2006): Elektronische Übermittlung der Adress.Certified (AC) Geo Daten 2005. Marshall, J. M. (1974): Insurance Theory: Reserves versus Mutuality, Economic Inquiry 12, 476 – 492. Mueller, D. (2003): Public Choice III, Cambridge. Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 Flood Risk Exposure in Austria 613 Olson, M. (1982): The Rise and Decline of Nations, New Haven. Prettenthaler, F. / Hyll, W. / Türk, A. /Vetters, N. (2004A): Internationale Erfahrungen mit nationalen Risikotransfersystemen im Zusammenhang mit Großschadenereignissen aus Naturkatastrophen. Analyse der Hochwasserereignisse von 2002 – Flood Risk, Studie im Auftrag des Umweltbundesamts, Bundesministerium für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, Vienna. Prettenthaler, F. / Hyll, W. / Türk, A. / Vetters, N. (2004B): Finanzielle Bewältigung von Hochwasserschäden, Präsentation am 24. November 2004: Flood Risk – Analyse der Hochwasserereignisse vom August 2002, Strategien und Maßnahmen, Bundesministerium für Verkehr, Innovation und Technologie, Vienna. Prettenthaler, F. / Vetters, N. (2005): Finanzielle Bewältigung von Naturgefahren: Vorschläge zur Reform des österreichischen Modells. InTeReg Working Papers Series Nr. 21 –2005, JOANNEUM RESEARCH Forschungsgesellschaft mbH, Graz. Rudolf-Miklau, F. (Presentation on July 4 th 2006): Hochwasser 2005, Ereignisdokumentation – Bericht Österreich (Hydrographischer Dienst – Bundeswasserbauverwaltung – Wildbach- und Lawinenverbauung), Bregenz. Sandroni, A. / Sqintani, F. (2007): Overconfidence, Insurance, and Paternalism, American Economic Review 97, 1994 – 2004. Schwarz, K. (2002): Kapitalstockschätzung in der VGR, Statistische Nachrichten 2, 124 – 127. Schwarze, R./ Wagner, G. G. (2007): The Political Economy of Natural Disaster Insurance: Lessons from the Failure of a Proposed Compulsory Insurance Scheme in Germany, European Environment 17, 403 – 415. Shavell, S. (1979): On Moral Hazard and Insurance, Quarterly Journal of Economics 93, 541 – 562. Sinabell, F. (2004): Hochwasserschutzmaßnahmen aus ökonomischer Sicht, in: F. Sinabell / S. Trimmel, Ökonomische Analyse von schadensminderden Maßnahmen im Hochwasserschutz. Studie des Österreichischen Instituts für Wirtschaftsforschung im Auftrag des Umweltbundesamtes, Vienna, 1 –42. Sinabell, F. / Trimmel, S. (2004): Ökonomische Analyse von schadensminderden Maßnahmen im Hochwasserschutz, Studie des Österreichischen Instituts für Wirtschaftsforschung im Auftrag des Umweltbundesamtes, Vienna. Sinabell, F. / Url, T. (2006): Versicherungen als effizientes Mittel zur Risikotragung von Naturgefahren, Studie des Österreichischen Instituts für Wirtschaftsforschung im Auftrag des Verbands der Versicherungsunternehmen in Österreich, Vienna. Statistik Austria (2006): Auf dem Weg zu einem nachhaltigen Österreich, Indikatoren Bericht, Bericht im Auftrag des Bundesministeriums für Land- und Forstwirtschaft, Umwelt und Wasserwirtschaft, Vienna, Available at: http: // www.nachhaltigkeit.at / strategie / pdf / indikatorenbericht / IndikatorenBericht_2006_080606.pdf (downloaded on Feb. 19 th 2007). Statistik Austria (2001): Gebäudezählung, Statistik Austria, Vienna. Swiss Re (2006): Natural disasters and man-made disasters 2005, Sigma 2. Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34 614 Thomas Url and Franz Sinabell UBA (Umweltbundesamt) (June 2007): Data on Urban Land use Provided by Gebhard Banko, mimeo, Vienna. Winter, R. A. (2000): Optimal Insurance under Moral Hazard, in: Dionne, G., Handbook of Insurance, Dordrecht, 155 – 183. Schmollers Jahrbuch 128 (2008) 4 OPEN ACCESS | Licensed under | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.128.4.593 | Generated on 2023-01-16 13:35:34