Why we're so bad at dealing with unlikely disasters like the Louisiana oil spill.

Who's winning, who's losing, and why.
May 3 2010 9:33 PM

Oh Well

Why we're so bad at dealing with unlikely disasters like the Louisiana oil spill.

Every industrial disaster yields lessons, and the oil spill off the coast of Louisiana last month will be no different. BP should have had more safeguards against explosions, we will learn. Regulation should have been tighter. Drilling should have been banned so close to shore.

But the real lesson of the oil spill may be how bad we are at dealing with unlikely but disastrous events. "We deal with them by ignoring them until they happen, and then overreacting," says John Harrald, a professor at George Washington University's Institute for Crisis, Disaster, and Risk Management. It's one thing to look back and figure out how the oil spill could have been prevented. It's another to grapple with the combination of poor foresight and 20/20 hindsight that makes preventing these meltdowns so difficult.


The problem begins with the way we assess risk. The less likely an event, the less a company—or a government, or an individual—feels the need to guard against it. No one knows yet what exactly caused the BP spill. But it does appear to be an example of several systems failing at once. We know there was an explosion (Failure No. 1) that caused the whole oil rig to collapse (No. 2), at which point the blowout preventer—a valve designed to seal off the well near the ocean floor—did not activate properly (No. 3). The chances of one of these things happening are not insignificant. The odds of all three happening at once are vanishingly small.

"Look at the expected value of a blowout of a well," says Harrald. ("Expected value" of risk is a mathematical term that means probability multipled by consequences.) "It's all the way down there with natural seepage and road runoff. It's a meaningless value." Any cost-benefit calculation would therefore ignore or heavily discount the possibility—even if the consequences are monstrous.

Another problem is that some precautions make sense only in hindsight. Before 9/11 and the shoe bomber, for example, the precautions that the TSA now takes—locking cockpits, banning fluids, requiring that shoes be removed—would have seemed absurd. "It's hard to get the public's attention," says Greg Shaw, also at George Washington. "It's hard to inconvenience them based on the idea that this may happen." Likewise, the precautions necessary to prevent the BP oil spill—some are blaming the absence of a sonar device used on rigs in Norway and Brazil to shut off the flow of oil—would probably have been dismissed as overly draconian. Now we'd be crazy not to require it. "It becomes very emotional after the fact," says Matthew Kotchen, an professor of environmental economics at Yale University.

Then there's the difficulty of cost-benefit analysis when it comes to low-probability, high-consequence events. The full costs of an oil spill are difficult to estimate—even decades later. After the Exxon-Valdez spill, Exxon paid more than $3.8 billion for clean-up and damage, plus about $500 million in punitive damages. But some lawyers argue that the true cost of the spill is much greater. There are many different ways to assess environmental damages—you could calculate it as cost to local fisheries, harm to the tourism industry, or even how much the average American would pay to keep the ocean unpolluted. If a company wanted to conduct a cost-benefit analysis of whether to pay for a particular safeguard on each of its oil rigs, where would it start? Plus, a well isn't like a tanker; you can't always tell how much oil it contains. So calculating the costs of an accident would be that much more difficult. That doesn't mean oil companies shouldn't take precautions against spills. It's just hard to calculate the value of those seemingly excessive precautions ahead of time.

When it comes to actually fixing problems, we're almost as incompetent. "We tend to regulate for that very specific event," says Harrald. After the Exxon-Valdez spill, for example, Congress passed a law that banned that ship from traveling in Alaskan waters. (For branding reasons, Exxon renamed the ship the SeaRiver Mediterranean.) At the same time, certain ports in Alaska became highly regulated, while others in Seattle and San Francisco remained lax. It's as if Congress passed a law against lightning striking twice.

Likewise, the BP spill has provoked a fast response. The White House started backpedaling from its promises to lift the moratorium on offshore drilling. "No domestic drilling in new areas is going to go forward until there is an adequate review of what's happened here and of what is being proposed elsewhere," said White House senior adviser David Axelrod. Gov. Arnold Schwarzenegger rescinded his support for drilling off the coast of California. Interior Secretary Ken Salazar ordered inspections of all deep-water operations in the Gulf of Mexico.

But overly narrow regulation can have unintended consequences. A crackdown on off-shore drilling might be politically smart, but it could drive business to countries with even less regulation of their oil industries. It could also boost demand for tankers, which have historically spilled more oil than American oil rigs.

The best responses to unlikely events take a holistic look at the problem, says Harrald. Rare events "raise opportunities to look at whole systems and get the political will and public support to do something about it." In this case, that might mean not only examining the safety of tankers and oil rigs, but also reducing our dependence on oil in favor of renewable energy technologies like wind and solar. But that might be the least likely event of all.

Like Slate on Facebook. Follow us on Twitter.

Christopher Beam is a writer living in Beijing.

  Slate Plus
Slate Picks
Dec. 19 2014 4:15 PM What Happened at Slate This Week? Staff writer Lily Hay Newman shares what stories intrigued her at the magazine this week.