Risk Analysis, Vol. 26, No. 4, 2006 DOI: 10.1111/j.1539-6924.2006.00790.x Direct Estimation of the Cost Effectiveness of Tornado Shelters Kevin M. Simmons 1 and Daniel Sutter 2* This article estimates the cost effectiveness of tornado shelters using the annual probability of a tornado and new data on fatalities per building struck by a tornado. This approach differs from recent estimates of the cost effectiveness of tornado shelters in Reference 1 that use historical casualties. Historical casualties combine both tornado risk and resident action. If residents of tornado-prone states take greater precautions, observed fatalities might not be much higher than in states with lower risk. Estimation using the tornado probability avoids this potential bias. Despite the very different method used, the estimates are $68 million in permanent homes and $6.0 in mobile homes in Oklahoma using a 3% real discount rate, within about 10% of estimates based on historical fatalities. The findings suggest that shelters provide cost-effective protection for mobile homes in the most tornado-prone states but not for permanent homes. KEY WORDS: Benefit-cost analysis; natural hazards; risk mitigation; tornadoes 1. INTRODUCTION Tornadoes are nature’s most violent storm and have killed over 6,400 Americans since 1940. Over the past several decades, wind engineers have designed a new generation of below-ground shelters and above- ground saferooms capable of protecting residents from even the most powerful tornadic winds. Tornado shelters have received considerable attention from the news media and policymakers in recent years, be- ginning with a segment on research at the Wind En- gineering Research Center at Texas Tech University on NBC’s Dateline after the Jarrell, Texas tornado in 1997. The Federal Emergency Management Agency (FEMA) decided in 1998 to publicize tornado shel- ters as part of its National Mitigation Strategy and re- 1 Department of Economics and Business Administration, Austin College, Sherman, TX, USA. 2 Department of Economics and Finance, University of Texas–Pan American, Edinburg, TX, USA. ∗ Address correspondence to Daniel Sutter, Department of Economics and Finance, University of Texas–Pan American, Edinburg, TX 78541-2999, USA; tel: 956-381-3354. leased the first edition of its publication Taking Shelter from the Storm. Following the May 3, 1999 Oklahoma City tornado, FEMA and the State of Oklahoma col- laborated on the Oklahoma Saferoom Initiative, with rebates of up to $2,000 for the installation of a tornado shelter. In 2003, Du Pont introduced a saferoom prod- uct made of kevlar and promoted in an advertising campaign featuring NASCAR star Jeff Gordon. The interest in tornado shelters naturally raises the question of their cost effectiveness as a safety measure. Individuals and policymakers will want to know whether tornado shelters offer good value rela- tive to investments to reduce other threats to life and limb. Recently, Merrell, Simmons and Sutter (1) esti- mated a cost per life saved for tornado shelters based on average tornado fatalities and a regression model of tornado casualties for Oklahoma for 1950–1999. A regression model controls for changes in the qual- ity of tornado warnings or population density over time and thus provides a better estimate of current expected fatalities. The cost per life saved for tornado shelters based on average annual fatalities and a 3% 945 0272-4332/06/0100-0945$22.00/1 C  2006 Society for Risk Analysis