April is the peak of the Southeast spring tornado season. Alabama avoided any major tornadoes, but two events in Pike County illustrate the challenges of forecasting and how tornadoes that never happen can be costly.
Last Thursday, a tornado struck unexpectedly northeast of Troy. Earlier on April 5, a feared significant outbreak did not materialize. The impact on Pike County of the April 5 non-event exceeded last week’s tornado.
Both events illustrate limits of meteorologists’ understanding of tornadoes. Conditions on April 5 were favorable for violent (rated EF-4 or EF-5 on the Enhanced Fujita Scale) tornadoes. The Storm Prediction Center had much of Alabama under a moderate risk (the second-highest risk level), with parts of Georgia and South Carolina under a high risk. No severe weather watches were in effect for last week’s tornado.
First let’s consider the surprise tornado. Spin-up tornadoes in thunderstorms and along squall lines are difficult to forecast or even detect. Such tornadoes occur in the bottom portion of the storm, basically below the radar. Weather radars more readily observe rotating supercell thunderstorms.
Consider next the non-event from April 5. Exactly when supercell thunderstorms spawn tornadoes is not yet known. Despite favorable atmospheric conditions, Alabama experienced no strong tornadoes.
Last week’s spin-up tornado almost surely had less impact on Pike County than the April 5 non-event. Why? Partly because spin-up tornadoes are almost always rated EF-0 or EF-1 on the Enhanced Fujita Scale. Thursday’s tornado was rated EF-1 with winds estimated at 95 mph. The impacts of weak tornadoes are similar to thunderstorm winds. Well-built buildings will not be blown down, but falling trees can crash on homes or cars with lethal force. Unsecured mobile homes can also be blown over. The measures which protect people against thunderstorm winds generally keep people safe in spin-up tornadoes.
The costs of readying for a significant tornado outbreak, by contrast, are substantial. Let’s consider just one impact, from school cancellations. When children stay home from school, parents often have to as well. The lost work and production is costly.
How costly? Suppose that one adult misses work for every ten students who stay home from school. Average earnings in Alabama are $22 per hour, but some production lost when an employee is absent can be made up, say by working from home, adjusting schedules, and so forth. If $10 per hour of production can’t be made up, the cost of a missed work day is $80, or $40,000 in Pike County with our 5,000 K-12 students.
Missed classes, at Troy University in addition to the grade schools, are trickier to value. Tuition or state spending per student-day would overstate the loss, since only one day out of a 180 day school year or 75 day college semester was lost. Lessons could be adjusted so that only the least significant material would be missed; the impact would be as if the term were one day shorter. Even a conservative estimate of the value of the marginal school day produces a value in excess of $100,000. So the education-related impact in Pike County on April 5 was around $150,000. With schools closed across Alabama, Georgia and South Carolina, the total cost of this forecasted outbreak surely was in the tens of millions of dollars.
This is not meant as criticism of school officials or tornado forecasters. Knowledge is not perfect, and the prudent decision when facing risk will necessarily appear costly when the feared event does not materialize.
Considering such costs informs us about the value of improved tornado forecasts. Spin-up tornadoes are not very costly, so better warnings here would not be very valuable. The cost of preparing for a tornado outbreak is large and avoidable with better forecasts, so research here would yield significant value to society.
Life is about decisions, and economics helps identify the consequences of these decisions. Sometimes we find surprising results, like how a tornado which didn’t occur can cost more than an actual tornado. And surprising findings help lead us to better decisions.
Daniel Sutter is the Charles G. Koch Professor of Economics with the Manuel H. Johnson Center for Political Economy at Troy University and host of Econversations on TrojanVision. The opinions expressed in this column are the author’s and do not necessarily reflect the views of Troy University.