The Economics of Natural and Unnatural Disasters

6 - The Socioeconomic Impact of Tornadoes

103 6 The Socioeconomic Impact of Tornadoes Daniel Sutter University of Texas–Pan American Kevin M. Simmons Austin College Tornadoes are nature’s most powerful and destructive storms, capable of producing winds in excess of 300 miles per hour, yet they are notoriously capricious, leveling one home and leaving the next undamaged . The United States experiences more than 1,200 tornadoes per year, and since 1900 over 15,000 lives have been lost in tornadoes. The deadliest tornado in U.S. history, the 1925 Tri-State Tornado, tracked across three states and killed 695 persons, devastating entire towns. Tornadoes have occupied a place in the national consciousness at least since the 1939 movie The Wizard of Oz, when a Kansas twister blew Dorothy and Toto to Oz. Every spring thousands of people spend weeks trekking across the Plains chasing tornadoes. How can economists or social scientists contribute to our knowledge of tornadoes? While cloud dynamics and the technical properties of weather radars are outside these fields, economics can help us understand the impact of tornadoes on society. Economics can provide relevant evidence on several issues related to societal impacts: • Have tornadoes become less deadly over time? • If so, how much have the efforts of the National Weather Service (NWS) contributed to this? • What measures offer the greatest potential to reduce casualties in a cost-effective manner? 104 Sutter and Simmons An understanding of the causes is necessary to reduce the impacts of severe weather. Just as physicians must understand the causes of illness to successfully treat patients, meteorologists require information about societal impacts. Attempts to reduce casualties not founded on solid analysis could prove unsuccessful or incur excessive costs. Tornadoes also provide evidence on some questions of significance to policymakers: • People sometimes have difficulty making sense of small risks of death and either overestimate or underestimate these risks (Camerer and Kunreuther 1989; McClelland, Schulze, and Coursey 1993). Is misperception of risk a problem with tornadoes? • Can an economic model of information help us understand peoples’ reactions to hazard warnings? • How prevalent is underpreparation for natural hazards? Hurricane Katrina has raised the issue of poor societal preparation for hazards to high salience for policy (Meyer 2006). Because of the broad reach of tornadoes (they have occurred in all states), their impacts depend on the preparations and actions of essentially all Americans, a fact that underscores the importance of evidence regarding these events. This chapter analyzes the impact of tornadoes on the United States and is organized as follows. The next section reviews the aggregate impact of tornadoes on the nation, including three main components: 1) the cost of casualties, 2) the value of property damaged or destroyed, and 3) the cost of responding to tornado warnings. Overall the monetized cost of tornadoes is $4.6 billion per year. We then discuss findings on the determinants of tornado casualties, and we use these findings to analyze how the impacts might be reduced. The final section offers a brief conclusion. THE SOCIETAL COST OF TORNADOES Tornadoes threaten life and limb, and they damage and destroy property. Tornado warnings are also costly, because people must dis- The Socioeconomic Impact of Tornadoes 105 rupt their daily activities to take shelter during a tornado warning. To provide perspective on the impact of tornadoes, we monetize the value of casualties, damages, and sheltering costs, based on U.S. averages for 1996–2006. Damage is the easiest to monetize, and we use inflationadjusted property damage as reported by the NWS, which averaged $1.07 billion annually (in 2007 dollars).1 Note that 1996–2006 included the tornado with the greatest reported damage in U.S. history, the May 3, 1999, Oklahoma City F5 tornado.2 A total of 645 tornado fatalities occurred between 1996 and 2006, or 58.6 per year. Comparing fatalities with damage requires application of a dollar figure for the lives lost. The value of a statistical life as revealed in market trade-offs constitutes a reasonable way to value lives for such public policy purposes.3 We use the value of a statistical life applied by the Environmental Protection Agency in a benefit-cost analysis of the Clean Air Act (EPA 1997). The EPA used a figure of $4.8 million in 1990 dollars, based on a meta-analysis of dozens of published studies. Adjusting this value for inflation yields a value of $7.6 million in 2007 dollars. The monetized value...