Reinsurance provides a significant and essential form of risk-spreading capacity for primary insurers. For natural catastrophes, this risk-spreading normally takes the form of an "excess" contract; primary insurers retain the first tier of losses up to a "trigger point" above which they purchase reinsurance, which operates up to a specified "exit point" or upper limit. After the catastrophes of the past 2 decades, reinsurers are leaving more of the risks with primary insurers, by increasing trigger points and lowering exit points (Stipp, 1997).
Many of the vulnerabilities experienced by primary insurers also apply to reinsurers. Several reinsurers became insolvent or were absorbed by larger firms during the crisis period of 1989-1993 (ISO, 1999; Mooney, 2000). The unexpectedly costly European windstorms of 1999 caused further problems (Andrews, 2000). For example, an already weakened Australian reinsurer covering these storms became insolvent despite total assets of US$2.3 billion (Howard, 2000a). According to the Insurance Information Institute (III, 2000a), the world's catastrophe reinsurance industry " lacks the capacity to insure mega-losses in excess of US$50 billion." Government reinsurance systems also have shown signs of stressas evidenced in France, where reserves fell by 50% during the 1990s and reinsurance rates rose sharply (CCR, 1999).
An additional source of vulnerability arises from regulatory uncertainties, such as the degree of flexibility afforded in withdrawing from markets and risks and in raising insurance prices (Davidson, 1996; Insurance Regulator, 1998; III, 2000a; Ryland, 2000). In some jurisdictions, regulators have restricted policy cancellations and nonrenewals following natural disaster losses such as Hurricane Andrew (ISO, 1994a,b; Lecomte and Gahagan, 1998). Recent requests from Florida insurers to double rates to protect insurers from hurricane risks also have been resisted by regulators (III, 2000b). On the other hand, under some conditions regulators can force insurers to withdraw from markets or otherwise change their business practices so they maintain minimum solvency requirements (GAO, 2000a). Pre-event accumulation and taxation of reserves also is an important issue, and policies vary by country (Eley, 1996; Davidson, 1997).
Under climate change, sustained increases in the frequency and/or intensity of extreme weather events could stress the government sector itself as a provider of insurance, a provider of domestic and international disaster preparedness/recovery services, and an entity that itself manages property and undertakes weather-sensitive activities (e.g., ranging from mail delivery to operation of military facilities near coastlines or waterways). Increasingly, governments seek to cap or reduce existing exposures (ISO 1994b, 1999; Gastel, 1999; Pullen 1999b; FEMA, 2000; III, 2000b). Governments in developing countries participate especially deeply in weather-related risks, given the low level of private insurance availability and often a higher level of government-owned infrastructure.
Disaster relief provided by the U.S. government has totaled $30 billion since 1953 (Changnon and Easterling 2000). Nearly half of these losses have occurred since 1990, and inflation-corrected payments rose six-fold between the late 1960s and the early 1990s (Easterling et al., 2000a). These costs do not include temporary housing, unemployment insurance, and small business loans also provided by government.
Governments are particularly sensitive to changes in flood- and crop-related losses because they often are the primary or sole providers of such insurance, and climate changes are expected to exacerbate these losses (see Chapters 4 and 5; Rosenzweig et al., 2000). U.S. government-insured crop/hail losses grew 11-fold between the 1950s and the 1990s (Easterling et al., 2000a). In Japan, the majority of international relief7-8 billion yen in 1990is related to floods (Sudo et al., 2000). Solvency is a material issue for government programs, as exemplified by the $810 million deficit in the U.S. flood insurance program in the mid-1990s (Anderson, 2000). U.S. crop and flood insurance programs have never been profitable (GAO, 2000a; Heinz Center, 2000). The French catastrophe reinsurance fund (Caisse Centrale de Réassurance) had become depleted as of the late 1990s and could no longer deal with a major catastrophe from accumulated surplus (CCR, 1999).
Other reports in this collection