Smoke on the Water

It’s hard to figure out what to say about BP that’s not either political or a joke. In fact, it’s hard to not view BP as a joke right now and to view how they’re handling the oil spill as a sort of Royal Dutch Shell Game.

Okay, sorry. I’m going to try to not tell too many oil company jokes here 🙂

Leaving aside all the political furor over the oil spill, it’s interesting to look at this from a risk management perspective. Psychologically, we view risk in ways that do not always conform the reality of the risk. For example, we tend to assume that if an event occurs once at a certain point in time, then it’ll happen again at that point in time: after the October 1987 market crash, for years afterward some people would jump out of the market in October just in case. Why? Well, it happened once in October (twice if you count the 1929 crash!). In fact, the 1987 crash was a unique event.

Similarly, when a disaster occurs and then doesn’t repeat, eventually we tend to assume not only that it won’t repeat but that the odds of its repeating are going down the longer the time that has passed. Thus, we had a major oil disaster in the Gulf in 1979. We had various smaller spills, but nothing so large after that (the Exxon-Valdez being a somewhat different kettle of oily fish). The natural assumption becomes that the problem won’t recur, either because we know more or because our technology is better or because the stars are in the correct alignment, or whatever reasons we might assign to the (non-)event.

Now, BP did have financial incentives to make these assumptions, and I quite believe those incentives made a big difference to their planning. This left them wide open to what Nassem Taleb calls a Black Swan: an event whose probability cannot be calculated but whose effects are potentially catastrophic (or beneficial, though not in this case). Part of what made the event so powerful is that BP was so manifestly unprepared to deal with it.

I suspect that part of the non-financial reasoning behind BP’s decision is that we no longer really think about oil as dangerous. Sure, we *know* it’s dangerous, but it’s been a part of our lives for so long now that it doesn’t have emotional impact any more. Compare oil to, say, nuclear energy. How did you react? Which is more dangerous? Which can do more health damage? How do you know?

Most people I’ve spoken to have a much different reaction to nuclear energy than to oil. Why? We don’t think about oil, while we do think about nuclear energy. It’s been a while, but my recollection of Three Mile Island is that it affected a much smaller geographic area and many fewer people than the oil spill. Again, this represents a misapprehension or confusion in how we assess risk. Nuclear energy is *scary* and oil is familiar.Fact is, if you have oil heat and your tank ruptures, you’ve got a major problem on  your hands. In some cases, it can be a real problem when familiarity breeds contempt.

Part of risk assessment needs to be not just how likely is an event to occur, which cannot always be calculated, but what gives us reason to believe that it won’t occur? As we see, just because it hasn’t happened recently doesn’t mean it won’t. On the flip side, some events really do work that way: the longer it takes to repeat, the longer it is likely to take. We thus need to understand which type of event we’re dealing with, otherwise we spend our time and energy defending against events that really are highly improbable or unlikely to do much harm while leaving ourselves wide open to the events that have a constant probability no matter how much time has passed and which can do devastating damage.

Leave a comment

Your comment