With regard to Mark Thoma's June 17th post, "The Enron Loophole":
Paul Krugman has asked many times, if speculation is pushing up oil prices, why is there no evidence he can find of hoarding, of increased inventories? (see here for a good example)
Some have argued that the records of inventories are not that accurate; that it's still possible to hide a substantial amount of hoarded oil. I have little expertise in oil inventory recording, so I don't want to offer an opinion.
But a point I would like to add is that if the supply and demand curves for oil in the short run (or very short run) are very inelastic, then you wouldn't have to hoard much to push the price of oil up a lot. And if you only have to hoard a relatively small amount of oil, then it might not be that hard to have it not detected in the inventory records.
Imagine (or draw) supply and demand curves (like in Krugman's graph) that are almost straight lines (Again, I'm thinking in the short run, or very short run. In the long run, the demand curve at least is very elastic). In that case, if you move the quantity that gets to the market back just a little bit, by putting it in storage, the price jumps a lot.