Catastrophic events present special challenges for economics and risk management since they have an immediate impact on a wide range of stakeholders, can have severe long-term economic and social consequences, and are difficult to assess quantitatively. As these events normally have a low probability of occurrence, there are limited historical data on which to base estimates of the risks, and there is considerable uncertainty associated with experts’ risk assessments. The terrorist attacks of September 11, 2001 are the most costly disaster in the history of insurance and have led both insurers and reinsurers to reevaluate under what conditions they can provide coverage against this risk. We examine the conditions for insurability of risks and conclude that terrorism presents special problems due to the dynamic uncertainty of the risk, information-related issues as well as correlated catastrophic losses between different lines of coverage and government influencing the risk through foreign policy and national security. These factors may explain the unwillingness of private insurers in the United States to offer coverage alone following 9/11 and why the US Congress passed legislation that provides government protection against catastrophic terrorist losses. We argue that the special characteristics of terrorism call for government participation in any terrorism insurance program to be based on public-private partnerships. This need has been recognized by most countries through the creation of national temporary programs based on public-private partnerships for covering (mega)-terrorism. Since the creation of these programs, the level of demand for non-compulsory commercial terrorism coverage has remained low in countries such as the U.S. and Germany. The paper discusses some factors that explain this behavior. If the low level of demand continues, a large-scale terrorist attack will likely have a more devastating effect on business and social continuity today than after 9/11 because losses will not be diversified in the national and international insurance and reinsurance industry. This raises the question as to whether terrorism insurance should be mandatory and, if so, how would such a program be administered? The report also discusses the most recent developments in quantitative risk modeling. A wide range of stakeholders are likely to find these models useful for evaluating their exposure through alternative scenarios but they currently are not able to predict the likelihood of specific terrorist actions. We conclude that better data are needed to evaluate alternative public-private partnerships for encouraging risk reduction measures and providing sustainable programs of insurance against terrorism should new large-scale attacks occur...