Watching the Fundamentals
Not many years ago, you could gauge the price and direction of oil and petroleum products in the US just by looking at the levels and trends of reported inventories. Anyone doubting that this no longer works, and that the US fuels market is strongly connected with the global market, need only look at the current market fundamentals as reflected in the inventories.

The Energy Information Agency of the Department of Energy released their latest weekly report on oil and oil products last Friday, showing production, imports, refinery utilization and inventories. The results are consistent with the recent pattern, with US crude oil inventories slipping but still at the top of their seasonally-adjusted historical-average range. Gasoline stocks exhibit a similar picture, both nationally and regionally, except for some tightness in the Rocky Mountains region. Distillate stocks (diesel and heating oil) are climbing toward the top of their seasonally adjusted range, as you'd expect. Anyone who'd spent the last several years on a desert island and was shown these inventory trends might reasonably think that crude oil was at about $25 and falling, with gasoline no more than a buck and a half at the pump.

So how can you have these kinds of market fundamentals and still be stuck at $60+ West Texas Intermediate crude and $3.00 gasoline? The details are complex, but I think two factors stand out. First, the fraction of crude oil we must import continues to rise, accounting for more than 10 million barrels per day (MBD) of the nearly 16 MBD run in US refineries. Imports of refined products also now make up about 10% of the 20+ MBD of gasoline, diesel and other petroleum products sold in the US. As long as the factors supporting the international prices of these commodities persist--high demand, tight refining capacity, and almost no spare crude production capacity--prices can't slip much here, no matter how high our inventories get.

Then look at the way that sustained demand growth shifts the relationship of historical data, when you compare inventory levels in terms of the number of days of demand they represent. With US demand rising 1 or 2% each year, after a few years a high absolute inventory figure can look average, or even low, in terms of days' usage. For example, US gasoline inventory reports for July 2005 averaged 212 million barrels, 6 million barrels higher than for the same month five years ago. However, the 2000 figure represented 24 days' demand, compared to 2005's 22.3 days. Effective inventory is 7% lower, despite totaling more barrels, because we are driving more.

All in all, then, while tracking of the EIA's raw statistics has suggested for some time that prices ought to be heading lower, I wouldn't bet on it until we see demand in China or the US slowing , or some sizable chunks of new production coming on line. We may have to wait until next year for either of these to have an impact.