Despite the frequency and expenses of natural disasters, there exists no system in either the public or private sector for consistently compiling information about their economic impacts. The committee was asked to provide advice regarding which data should be consistently used in compiling estimates of the losses of natural disasters.

The committee recommends that any data collection effort focus on the losses as a result of natural disasters, or negative economic impacts. The loss from a disaster is a broader concept than its cost, a term that conventionally refers only to the losses that are reimbursed by insurance companies and governments.

We suggest that the following specific types of data be assembled according to the following:

• One agency of the federal government should be made responsible for compiling a comprehensive data base containing the losses of natural disasters, adhering to the structure outlined in Table 2–2 (see page 29) wherever it is feasible. The committee believes that the Bureau of Economic Analysis (BEA) within the U.S. Department of Commerce, in consultation with FEMA and other federal agencies involved in natural disaster preparedness, response, and mitigation activities, is best suited for this purpose.
• The types of direct loss data that should be included in these loss estimates are listed in Table 2–2. The loss data should be classified according to: (1) who initially bears the loss: government, insurers, businesses, individuals, or nongovernmental organizations (NGOs); and (2) the type of loss: property, agricultural products, human losses, cleanup and response costs, and adjustment costs. Distribution of Table 2–2 to parties affected by disasters for comment may produce additional items for which data should be sought and compiled.
• These estimates should be made with the input of the following groups: Property Claims Services (PCS), the Institute for Business and Home Safety (IBHS), and the National Association of Insurance Commissioners.
• Direct losses include quantified (and, where appropriate, monetized) losses to human life, property, agricultural products, cleanup and response costs,

• and adjustment costs, such as temporary living aid. Special efforts must be made to assemble data on losses of uninsured individuals and businesses, a major gap in existing loss data.
• The Office of Management and Budget (OMB), in consultation with FEMA and other federal agencies involved in natural disaster preparedness, response, and mitigation activities, should develop annual, comprehensive estimates of the payouts for disaster losses incurred by federal agencies. This will allow the federal government to better track federal disaster spending, which in turn should assist policymakers to better plan for and budget disaster-related expenditures.
• Indirect losses of natural disasters, such as temporary unemployment or business disruptions resulting from physical damage, are currently diffuse and rarely quantified. Efforts should be made to regularly collect data required for estimating these losses.
• Efforts in the modeling of indirect losses should be strengthened. A microsimulation model for forecasting the timing of business closures and indirect losses should be developed. In order to provide baseline data to calibrate this model, a survey of businesses should be conducted to understand their likely responses and abilities to cope with disasters. In addition, formal models for simulating business and economic restoration following disasters should be developed.

A comprehensive, reliable data base on the losses caused by disasters would be valuable to several groups. Perhaps most important, a baseline set of loss data, together with cost and benefit estimates of alternative mitigation measures, would allow the federal government and individuals and firms in the private sector to design and implement cost-effective strategies for mitigating the losses from natural disasters. Insurers could certainly use the data to improve their estimates of future payouts associated with disasters. And researchers and experts in disaster loss estimation could benefit from a standardized data base that would enable them to improve estimates of both the direct and indirect losses of disasters. These improvements in turn would assist policymakers in their efforts to devise policies to reduce the losses caused by future disasters.

Beyond providing an indicator of total natural disaster losses to the nation, the framework for loss estimation described in this report would also provide detailed information on losses. A better understanding of issues such as who bears disaster losses, what are the main types of damages in different disasters, and how those losses differ spatially, are of critical importance in making decisions about allocating resources for mitigation, research, and response.