Wednesday, December 16, 2009
This morning, I've been looking at internals of oil demand and their relationship to gasoline (after looking at overall demand a couple of days ago). Firstly, here's a reminder of what fraction of US product supplied goes in different kinds of products. This is the average of the weekly numbers for 2009:
Gasoline is of course the largest component, accounting for a little under half. So I applied the same seasonal adjustment methodology I used here), and applied it to the weekly gasoline supplied numbers. The seasonal adjustment factors you get from doing that are as follows:
That looks pretty sensible - Americans drive more and more through the spring and summer, then less in the fall except for around Thanksgiving and Christmas, and then least in Jan/Feb.
The seasonally adjusted gasoline numbers then look like this:
Note the 2001 recession! Gasoline demand hardly declined at all.. Again, gasoline demand does now seem to be recovering following the most recent recession.
Finally, here is the relationship with unemployment. Again, surprisingly weak, especially on account of the non-response of gasoline demand to the 2001 recession. (I had to go back and check the VMT trend in an old Oil Drum piece of mine to confirm the data look the same and I wasn't making some gross error). I guess people just borrowed their way through and kept driving?