[pup55]

I probably should have given this a thread of its own, since it is useful information that we might want to keep for future reference:




I went to the NOAA website of historical hurricane tracks:

NOAA historical hurricane tracks

and got a list of all hurricanes, category 1 or higher, that passed within 150 nm of Louisiana. In the last 10 years, there have been 10 of these storms, so an average of one per year.

I looked at the WTI price before and after the maximum wind speed. In the table above, "pre-storm" is the price 10 days before the event. The "storm run up" is the average price of the 5 days before the storm, and "storm recovery" is the average of the five days after the storm. Post storm is the price 10 trading days after the storm.

Also, I looked up the US crude oil production, a week or two before the storm, and at the low point after the storm.

So we learned a few things from this:

a. In a pretty substantial number of cases, about half the time, the post-hurricane price was about the same or lower than the pre-hurricane price.

b. In only about half of the cases was the "panic price" in the storm runup period higher than the immediate post-storm recovery.

c. There were 5 storms that at one point in the gulf were category 4 or higher, and the minimum effect on oil production was .7 mbpd. This might or might not seem like much, but nowadays this would be nearly 15% of US production. I combined Cindy and Dennis together, since they hit so close to one another, into one event.

d. Naturally this points out the really unusual nature of 2005, with three major storms in this area in one year.

e. Past performance is not necessarily a predictor of future results, but I believe we have already seen the worst case scenario, and lived through it, so I think we have an idea of what to expect. On average, we can expect one storm going through this area at some time during the year.

f. We will probably look back at Ivan being a really unusually destructive storm. It caused a 13% price increase, knocked out a lot of production, and killed thuderhorse.



Edit: changed the price calculation: price change = post-storm price minus pre-storm price divided by pre-storm price.

[pup55]

I did a little more checking into the interesting case of Hurricane Lili, which hit Louisiana in October of 2002.

Despite the fact that this was a category 4 hurricane, and it impacted basically the heart of the offshore stuff, and despite the fact that it caused nearly a 20% decline in US crude oil production for awhile, the WTI price barely budged.

Furthermore, the inventory situation was that there were only 270 million barrels of crude oil in inventory which at the level of consumption at the time, was only 19 days' usage. Today, we have 21 days usage in inventory, so even the inventory situation was a lot worse.

So, why did the markets not go crazy?

a. One theory: When we attacked Iraq in 2003, we took something on the order of 2 mbpd out of the world market. There was slack in the market in 2002, and there was none in 2004.

b. China demand really started to ramp up in 2003 and became a factor in world demand in 2004

c. All of this points to the fact that the overall supply did not grow as fast as the demand, so the stresses on the system by the time 2004 rolled around caused the market to be more vulnerable to these fear situations. You may call this PO if you want, it comes to the same.

d. You can make the argument that a similar hurricane, hitting in a similar place, today, would have an even higher effect. We are a couple more years down the line, demand is a couple percent higher, you never know what will happen.

e. The next question: Why did the prices not go even crazier during the Katrina and Rita storms of 2005? Answer? Maybe the Dennis and Emily events just prior to this sensitized the market already. If you take all of these storms as one big event, and look at the price change between mid-July, before Cindy hit, and ending with the panic just before Katrina, the overall pricing increased maybe by 15%. The effect was temporary, though, despite continued hurricanes that year, the market tamed down by the middle of October.

f. So, despite the overall pretty good inventory situation we are in, it would not be unusual to expect a similar hurricane today to cause a price increase of 15 to 20% over and above the prevailing price at the beginning of the event. This would put the price around $85 or so, for those keeping track. The market is overly sensitized from all of the supply issues this summer, so there is no way to tell (yet) whether this estimate is valid or not.

The odds are pretty good that we will have a chance to find out, sooner or later. We can refer back to this thread and see how good of a job we did of estimating the vulnerability in the system.

[AirlinePilot]

Your "slack in the market " in point a. above is far more important right now than it has been in the past. I see this as a major issue now due to the the markets being a bit more savvy than they were in the past about production issues moving forward.

If you dont have the confidence in OPEC's spigot anymore the premium for disruption is going to be higher. Katrina may have been an ugly case but my belief is that a CAT 3 or 4 moving up through the Houston ship channel would spell major problems for us, possibly for a very long time. I really think so far we have been LUCKY in this regard.