11 April 2005

Risk management

"The [oil] industry is risk averse," [Premier Danny] Williams said.

That's from a scrum the Premier gave on Sunday about the oil companies and the pending launch of an American satellite launch that will see debris from the rocket used landing  - at the closest -  25 kilometres away from the Hibernia and Terra Nova oil production platforms. He's been quoted in a number of news stories saying exactly those words.

Oil companies aren'’t actually risk averse. They are risk managers. Oil companies follow a “"discipline for living with the possibility that future events may cause adverse effects."”


They manage risks: that is, they calculate possible outcomes from certain decisions as well as probable outcomes - the likelihood of certain results - and then decide what to do on that basis. Many factors might be involved and decisions are based on historical records -  that is what has and has not happened  - as well as physical factors like geology, financial costs, available resources like cash and people and equipment, and how each of those has and might perform. They make projections: what will likely happen if we do this or that. They may even generate a range of possibilities. They also figure out what changes they can make to their own procedures to lessen whatever risks they identify. That's called mitigation.

We are all risk managers to one extent or another. The only truly risk averse people are those who will accept no possibility of anything bad ever happening. Truly risk averse people are usually found either in a mental institution or under medical care (and strong medication) since you simply can’'t exist on the planet without accepting some risk.

If you look back and read those words, you will see a lot of conditional language "“may"” or “"might"”. That’'s easy to understand.  No one can predict the future with a perfect level of accuracy. That’'s the reason why Danny Williams'’ standard on the Titan rocket launch is a level of risk aversion which is impossible to achieve. No one can predict with absolute certainty that such an event will not occur.

There'’s more to this than just philosophy or outlook.

Danny Williams'’ position is the one usually taken by lawyers for the plaintiff in a personal injury suit: no risk is acceptable. If there is a risk, then someone other the lawyer'’s client is going to write a cheque as compensation.

Danny Williams'’ concern at the outset of this story was understandable. He didn'’t have much information and what he did have didn'’t look good. He prudently sought more, accurate information. He also expressed legitimate concern about the level of risk to lives and objects. In most cases, one would expect that as Danny Williams gained more information on this issue, he would have realized that the likelihood of a catastrophe was about as small as it could be. In fact, the odds of anything happening that will affect the oil fields is really tiny.

It simply isn'’t very likely at all that either the rocket or any piece of the rocket will come close to the Hibernia and Terra Nova platforms, let alone strike them.

The Premier has complained about not having accurate predictions of risk. But look at the numbers he does have from people who know what they are talking about: one in 10 million or one in a trillion or even one in 10 trillion. As the Premier has learned more, no one has increased the likelihood of what he fears. Rather, they have refined their prediction of exactly how unlikely it is. He should be comforted but he isn'’t.

The question is why.

Most of the information the Premier needs to make a reasonable decision is openly available on the Internet. Within five minutes -– literally five minutes –- of hearing this story on Thursday, I knew the type of rocket, the payload and other details. Within an hour, I had found history on the Titan 4B rockets including official investigation reports into the two major crashes of these rockets. Not only that but I knew that the US Air Force had not only identified the causes of those crashes but they had taken action to eliminate the chance of those causes happening again.

I also knew, from my previous research going back two decades, that liquid fueled rockets like this one are very predictable. Point it somewhere and you can accurately predict where it is going to land. Titans started out as nuclear-tipped ballistic missiles. The Americans could put this thing down on a target within a few hundred feet of where they wanted it. By the same token, they could also predict where it wouldn’'t go with a similar level of confidence. Give it a spot, like an oil platform and the launch operators could do things to increase the chances of missing it completely.

Based on openly available information, I also know that every single launch has a self-destruct mechanism controlled by a human being. If the rocket goes outside the predicted performance and looks like it will become dangerous, one press of a button will end the whole launch in an explosion and a shower of debris. All that will happen well away from this province and the offshore fields. The 1998 accident Danny Williams holds up to justify his concerns ended just that way - deliberate destruction - a mere 4, 422 feet from the launch pad and 17, 000 feet in the air.

On the basis of all that information, I came to a conclusion that while there was a theoretical possibility of something happening, the likelihood of something from this rocket striking the platforms was so remote as to be as close to impossible as anything can get.

The oil companies have already made those sorts of calculations and with far more detail and expertise than I have. So have their insurance companies and in order to make those guys happy, the oil companies are actually will to increase the risks of other things happening to satisfy the insurers about the rocket. For example, with all the helicopter activity there is a measurable increase in the possibility and maybe even the probability a helicopter may crash. There is also a measurable risk the fields may be permanently shut down by this de-mobilization.

But it's their money. Oil companies can spend it as they wish based on their calculations. That'’s what they do: manage risk. The regulator -– the offshore board –- isn'’t going to interfere in this for one reason: if they order either a shutdown or a continuation of operations under the circumstances, the board and the governments it represents become liable in the off chance something bad happens.

Just go back and think about Danny Williams’' demand for an absolute guarantee of safety and you can see why his position is unreasonable. You can see why his ongoing demands for reassurance have long since passed the point of prudence. In order to meet his demand, the US either has to cancel the launch, pick a new trajectory or move the payload to California for launch. They aren'’t going to do either of those things.

Since it will go ahead, the only way to be absolutely safe is to be nowhere near it. Even moving the Terra Nova platform 50 miles to the west won'’t decrease the theoretical risk to zero. Given Danny Williams' standards, every person in Newfoundland Labrador is in mortal danger next Sunday. The 25 kilometres between the platforms location and the outer edge of the debris zone is as good as 250 ro 2, 500 kilometres away. Who can say the Titan won't take out a house in Mount Pearl, as easily as the Hibernia rig? From 150 miles up, the two places are not actually that far apart.

I think being in Mongolia would lessen the risk from the rocket, but it might raise other risks and drive up other costs. So why move it at all, except that the insurers are willing to accept a one in a quadrillion chance but not a one in a trillion chance?

And if we need an absolutely zero chance of bad things happening, then the Premier has a long tough road in front of him. He will have to bar oil tankers from our waters for risk of a catastrophic spill of oil. He will have to ban any planes from flying over the province or cars from driving in the province because the likely risk to life and property is much higher than the absolute zero he has set as the target.

He will have to ban not only firearms but also sharp-edged tools and potentially all edged tools from private possession in Newfoundland and Labrador. Live without the possibility of something bad happening? Live without butter knives. In every instance, the chance of something bad happening from cars, guns, planes and knives is actually much greater than the Titan rocket hitting the oil platforms or wiping out a house in King William Estates.

Look at it this way. The chance of any one spot in the debris zone being hit next Sunday from this one single launch is one in 10 million for the worst case scenario. The chance of any one person being killed in an airplane crash on any given day when they fly is one in 11 million. Since the platforms are outside the debris zone, the chances of anything hitting them is one in a trillion.

The chances of being killed in a car accident for that one person taking one car trip are one in 5, 000. The more driving you do or car rides you take in a given space of time and you increase the odds of having an accident. Ditto for the flying. Fly more often, change planes, travel longer distances and you add to the risk.

Draw you own conclusions on whether or not to fly to Outer Mongolia next Sunday rather than risk the Titan launch.

Think really carefully before taking a cab to the airport.

[This is a revised version of an earlier post to add new information.]