- Home
- Report n°3: Financial protection of critical infrastructure
Report n°3: Financial protection of critical infrastructure
-
Table of contents
- On the morning of September 11 2001 : financial protection and new vulnerabilities
- September 10, 2001 : terrorist risk in insurance policies
- On the morning of September 11 2001 : financial protection and new vulnerabilities
September 10, 2001 : terrorist risk in insurance policies
Before the large-scale attacks against the United States on September 11, 2001, coverage for terrorist activity was not a particular problem for the insurance industry in most industrialized countries, either because countries had already been exposed to such risk and had set up specific insurance schemes (France(11)), United Kingdom, Spain, Israel, South Africa, inter alia), or because other countries did not view terrorism as a potential cause of disaster, since they had never suffered it (United States).
The situation was particularly striking in the United States. Before the September 11 attacks, the terrorist risk was only rarely the subject of exclusion clauses in property and casualty insurance policies. Private markets were functional essentially because losses due to terrorist attacks in the United States were historically fairly low and non-correlated because perpetrators were terrorist groups acting independently for different causes(12).
Furthermore, in a post cold war context, the possibility that a country would be attacked within its own borders by alien terrorist agents was judged by most people to be an impossible scenario at that time (national American commission, 2004); as a consequence, terrorism was never excluded in standard insurance policies, and was therefore covered without any special reference nor payment of specific premiums.
A change in the nature of the risk
If insurers had paid more attention to their possible exposure to terrorist risk, they might have observed the radical change taking place in the last two decades as regards the actual nature of the terrorist threat. If that had been the case, it is very likely that they would have adjusted their liability long before the morning of September 11, 2001 (exclusions, limits on portfolio concentration, specific premium for terrorism risk).
To gain a better understanding of what was at stake by covering terrorist risk through the sole agency of insurance and reinsurance markets, one needs to observe in more detail the changes in the nature of international terrorism in recent years. It is not intended, quite obviously, to offer an extensive analysis of this evolution. Rather, this paper sets out to define some reference parameters, based on the most recent research on the subject. This should lead, in particular, to a better understanding of why the central issue discussed by the world of insurance today is no longer whether another attack of similar dimensions to September 11 is likely in the future, but when and where it will happen, and what with.
Walter Enders and Todd Sandler, two pioneers in political science studies of international terrorism, have recently published several works which clearly demonstrate that the nature of this risk had been profoundly modified in the last 20 years(13). And the paradox is striking. On the one hand, the total number of international terrorist attacks (not perpetrated by domestic interests) has dropped considerably on average between the 80s and the 90s. According to data available to the American State Department, 635 terrorist attacks were perpetrated in the world in 1985, 612 in 1986, 665 in 1987 and 605 in 1988 (U.S. Department of State, 2003). In comparison, 296 terrorist attacks were perpetrated world wide in 1996, 304 in 1997, 274 in 1998 and 355 in 2001, i.e. almost twice less on average than at the end of the 80s. This drop is largely due to the end of the East/West conflict which led to a significant fall in the number of terrorist groups perpetrating political attacks (Pillar, 2001 ; Chalk, Hoffman, and Kasupski, to be published).
On the other hand, although they are fewer in number, the attacks are no less devastating. There has been a radical evolution in these years graduating to larger scale attacks inflicting more human loss. In fact, observing the 15 most lethal events reveals that they all happened after 1982 (see Table 1 below). Furthermore, 80% of them were perpetrated in the last 10 years, between 1993 and 2004, including the large scale attacks in Madrid on March 11, 2004, which killed over 200 people and injured more than another thousand(14).
(11) Until 1986, acts of terrorism were a case for exclusion clauses in policies covering property damage insurance in France. Since a law dated September 9, 1986, this exclusion clause is deemed unwritten and can no longer be put forward by insurers: mandatory coverage clauses for acts of terrorism or attacks now prevail. Material loss is therefore always subject to compensation by insurance companies. Compensation for physical injury is insured by the Fonds de garantie des victimes d'actes de terrorisme et d'autres infractions (Fund for compensation to victims of acts of terrorism and other infractions) created in 1986.
(12) The first attack against the World Trade Center, which killed 6 people and injured a thousand more, was perpetrated in 1993 using a car bomb in one of the underground car parks. The attack caused 725 million dollars' worth of insured losses. The Oklahoma bomb attack in 1995, which killed 168 people, had been the largest ever perpetrated on American soil, but material losses were essentially to Federal property, and therefore covered by the Government (Swiss Re, 2002).
(13) See particularly Enders and Sandler (2004).
(14) For an in-depth international analysis of the subject of terrorism, see SOS Attentats (2003).
This radical change took place gradually over several years, with as a result a profound evolution in the very nature of the terrorist risk: at first more political in the 70s and 80s, then becoming religious and supported by extremist groups of the Al Qaeda type who now operate and publicly claim their capacity and determination to inflict particularly devastating attacks (Stern, 2002).
In other words, terrorism is no longer brandished only against a certain social order or for a certain political ideal, but more systematically against the "West" which is viewed by these religious groups as supporting other values that they consider to be divergent (Hoffman, 1998; Pillar, 2001; Wedgwood, 2002)(15).
In this kind of context, as Sandler and Enders recently wrote, "the events of 11 September with their massive casualties of innocent people of all ages came as no surprise to those of us who study terrorism and warned of an ominous changing nature of transnational terrorism." (Sandler et Enders, 2004). As the National Commission on Terrorist Attacks on the United States clearly pointed out, these warnings had not had sufficient impact by the evening of September 10, 2001.
(15) This is not meant as a detailed analysis of this evolution. Interested readers will find further material in the list of references.