Here are a couple of graphs for you. The first one is just the refinery margin, since 2004, compared to refinery utilization of the same week. You can see the two episodes of hurricane destruction, the one in 2004, Ivan, was actually quite serious. We all know what happened in 2005.
But you can see, there is only a rough correlation between refinery margin and factory utilization.
The second graph is just the same data as the first graph, only the refinery utilization is shifted 7 weeks to the left. You can see, all the maximum and minimum points, with the exception of the hurricane damage periods, line up almost perfectly.
I could probably go back and do some calculation or let MS-solver tell me exactly how long the lag is.
But, for the purposes of this exercise, suffice it to say that it looks like it takes about 7 weeks for any pricing maximum or minimum for the refiners to get their act together and do what the market is telling them to do. Hard telling why this lag is. Maybe they are scheduling their units on this kind of a time frame, maybe there is some other reason. These are big, complicated pieces of equipment, so I guess you cannot just flip a switch and walk away.
Based on this, I am ready to say that refinery utilization will actually drop for another several weeks, until awhile after the picking up of the margins. In fact, we should be able to calculate the approximate refinery utilization for the next several weeks based on the already existing refinery margin data.
Stupidly, there is probably some young engineer on the night shift of one of these refineries who knows this data, and also knows that there is a huge financial fattening up available to the refiner that anticipates this price increase and has a lot of inventory on hand to sell at just the moment of maximum pricing. But it does not seem to happen that way for some reason. Probably the reason is fear on the part of the management that they will be producing and sending product to inventory with expensive crude oil. I guess it is just human nature, and it takes about 7 weeks for someone to capitulate and finally get the message.
I am ready to prognosticate that the refinery utilization will bottom out at about 83.5% the week of December 7th. The refinery margin minimum was about 12 cents last week, and is 14 this week. Add 7 weeks to last week.
By then, however, the refinery margins will have turned around and be back in the 20-30 range. I am still not sure on exactly how this will happen, by some combination of crude oil price decrease, or gas price increase. With Thanksgiving coming up in three weeks, I am thinking get ready to pay more for gas.
Also, conceptually, the refinery margins could continue to go down next week, making last week a false signal. I cannot tell for sure yet.
The frequent viewers of PO.com are hereby treated to a little advance notice of the news story several weeks in advance, to make whatever preparations are needed.
Warning: I am just some guy on the internet. This post is for entertainment purposes only. Anyone who burns down their garage because they are keeping some quantity of gasoline in there on the basis of this post is on his or her own from a liability standpoint.