Friday, February 25, 2011

libya as a probability

If you've been following the news, then you are likely already familiar with the situation in Libya. It has been possible for people to get out of the country and many expatriates are getting out.

Libya, in some ways feels like Macondo to me. Superficially, this are hardly comparable events as one involves the change of power in a nation and the other a well control situation gone wrong. Deeper than that is the difficulty with truly robust emergency response plans. How do you properly plan for and respond to crises that have never been seen before? Related to this, but going still slightly deeper is how do you properly predict and plan for very low probability events?

By low probability, I only mean that on any given day or week or month these incidents (radical government change, catastrophic well control failure) are unlikely to occur. However, given enough time, activity, and repetition, these events actually become almost inevitable.

What happened at Macondo to Deepwater Horizon is our generation's Piper Alpha. Drill enough wells, take enough risks, skirt enough safety protocols and this will happen. I previously discussed this concept and still stand by the assertion that while Macondo was a low probability event, it was almost certain that it was going to happen eventually. It is only the exact when and where that cannot be foreseen.

Similarly, the continually slipping control of power for Gaddafi and its implications for Libya seem like they should have been viewed as inevitable. Obviously, hindsight makes this assertion seem easy but let's review the options. The removal from power of a 40+ year leader of a country creates a huge vacuum within a nation's structure and leadership. Regardless of this occurring due to natural causes, military coup, or popular uprising, the resulting loss of structure is what can be predicted. Yes, the nature of the change strongly dictates how the transition will go and what the end-state will be (democracy, military rule, chaos, protracted civil war etc), but at the very least short-term instability is almost certain. (Related question: What happens in North Korea, Cuba, Venezuela, or large portions of this list when power changes hands? Bonus question: What do the leaders of those countries think about what is happening in Libya?)

When a loss of structure occurs, the question is how prepared are we/you/others prepared for the change. If we can assert that this was inevitable and say so with a degree of confidence that is not strictly induced by hindsight, how could this have been planned for? Timing is crucial. However, timing is not something we can predict as accurately, though in the case of Libya, the precursors were in Tunisia and Egypt. Sure, some thought Gaddafi had tighter control and popular protests would not spill-over into Libya, but that's not important. Timing is still hard to predict and a robust emergency response plan needs to work regardless of timing because emergencies, by their very nature, are not planned and scheduled activities.

Timing, if known, would make things much easier. Planning for the unknown and an unknown time is difficult because it is really a question of resources and having them available when you need them. And when you need them is all the time if the timing is unknown and all the time is usually quite expensive. We cannot pretend to keep a helicopter/plane/boat on perpetual standby for evacuation that may never occur. This makes availability of resources directly related to robustness of a plan and brings the acknowledgment that all plans are limited and that adaptation is critical.

I am not proposing any specific emergency response plan since they need to be tailored to the circumstances relevant to your situation. Any such plan needs to be stress-tested and reviewed and updated to reflect the fact that plans get made, then the overall situation can shift and gradually change to a point where the plan no longer works. Perhaps no plan would have done perfectly in Libya, but this is why we adapt and learn and figure out a new plan. (Rather random science fiction analogy not likely to be gotten by most readers: The presence of the Mule in Asimov's Foundation series as an example of a disruptor of a very grand plan that then had to be steered back on course.)

Low probability now can often mean nearly guaranteed probability eventually. This is why we must plan.

No comments: