Crude oil prices have been driven up by demand, but is it demand for wet barrels or demand for futures contracts? If those are the only two choices, it must be speculative demand because there doesn't appear to be any shortage of wet barrels—either now or for delivery in the next year.

When President Bush went to Saudi Arabia in January 2008 to beg for higher crude oil production rates, they said, "No, there is no shortage of oil." They said they knew that because worldwide inventories were in a normal range. The logic of their position is that when consumption exceeds production, the difference can only be made up by drawing down existing inventories. The Saudis said that if they were to increase production, the incremental oil would just push up inventories and risk creating a price-crashing glut (which is, of course, what Bush really wanted).

It seems the Saudis are correct about inventories. Here is a plot of US crude oil inventories (excluding the Strategic Petroleum Reserve) from 1973 through February 2008 and the 12-month moving average. Nothing about recent inventory levels looks unusual. Data are from US Energy Information Agency.

Indeed, if one looks at inventories over shorter time periods, it seems that inventories have gone up during periods of higher prices (before the First Gulf War and since 2003). That suggests that prices have been driving inventory decisions rather than vice versa. In other words, prices didn't rise because inventories were shrinking; inventories increased because buyers expected the barrels would be priced higher later.

This morning I went looking for an expert analysis of the crude oil futures markets, how they may have changed in the last 5 years, and how such changes might affect prices. I found this piece by F. William Engdahl, whom I don't know and who may be, according to information at the end of the piece, someone with inordinate confidence in his own ability to see around corners. With that caveat, I recommend his piece for the insights it gives into the opaque world of crude oil futures and the argument that we are in a speculative bubble. He suggests that 60% of current prices (i.e., everything over about $45/bbl) results from demand for paper barrels, not wet barrels.

The contrary argument is that worldwide demand for oil has risen so far so rapidly, especially in China, that producers will soon be unable to keep up. Supporting this are one private communication and several press reports that the costs of increasing production in places like the San Joaquin Valley have risen so far so fast that it is not an attractive investment even at today's prices. However, Engdahl cites a recent Energy Information Agency report, which I have not looked at myself, saying that worldwide production capacity is continuing to grow and that by 2010 there will be 3-5 million barrels per day of production capability in excess of actual production.

If we aren't actually about to run short of oil, why do so many people say that? It is certainly possible that some of the talk is disinformation meant to make trading strategies pay off. There are also surely pundits who did straight-line extrapolations 2 years ago and haven't recalibrated their opinions to the recession and declining demand for gasoline in 2008. There are clearly those who know nothing about crude oil but are using the "peak oil" rhetoric to get government action for alternative energy sources, for drilling in ANWR, against global warming, etc., etc. And there are many, many bandwagoners who just repeat what they take to be the conventional wisdom.

In the end, I'm buying the speculative bubble hypothesis. I predict oil will trade below $100 before the end of 2008 and below $70 before the end of 2009. Meanwhile, the price could climb to $150 or more. Or it could fall off the cliff this afternoon and destroy another company like Enron before Memorial Day. Extreme volatility is the dark side of extreme liquidity.