How Much Higher Could Oil Go?
After months of watching oil prices edge higher and higher, it is increasingly difficult to imagine a return to lower prices, and correspondingly easier to forecast even higher prices. The Financial Times recently cited Goldman Sachs's suggestion of $105/barrel, as well as mentioning a French bank that thinks oil could go to $350 by 2015. Though the former seems at least plausible, I am skeptical of the latter, having watched similar estimates go down in flames in the past. But I'm also very curious about their assumptions. Those are usually the most interesting element of any forecast, rather than the final number.

What would have to happen for $350 to come true? First, you'd have to guess that the world economy (and especially China and India) will continue to grow at a healthy pace, dragging oil demand along with it. Another decade of 5% growth would expand global GDP by 60%, creating a richer world that might be less price sensitive about oil. The FT article indicated that Caisse D'Epargne sees a global oil supply shortfall of 8%. To make that more concrete, imagine going from today's 84 million barrels a day of demand (and supply) to 102 MBD of demand (2% growth for 10 years) but only 94 of supply. So their world of 2015 is effectively missing the equivalent of Saudi Arabia or Russia.

That would be fine as a disaster scenario, but it's hard to see how you'd ever just grow into a world like this. Supply and demand don't get that far out of whack without some serious substitution or other response that brings them back into line. That might take a few years, but ten? In ten years, with crash programs, the world could close the potential gap with a combination of efficiency savings and non-petroleum energy sources, including oil sands, gas-to-liquids, LNG, and maybe a bit of something more exotic. In fact one risk of this kind of effort is that it would be too successful, forcing oil prices below today's level and making the substitutes uneconomical. When you start looking at things this way, as a dynamic system, it's hard to see $350 oil--which would translate into US gasoline pump prices near $10/gallon--as a viable end state, rather than as a single snapshot out of a longer sequence.

Alternatively, could $350 simply be Goldman's $105 plus a massive dollar devaluation (to something like $4 to the Euro) and a switch from oil being denominated in dollars to the "basket of currencies" concept that OPEC has floated several times in the last couple of decades? The media is currently brimming with commentary from economists who are so concerned about the various US deficits that such an outcome might not surprise them. (I find this a much scarier scenario than the first one.)

In any case, although the last year has brought me around to the idea that high oil prices won't go away soon, I still remember those $100 oil predictions from thirty years ago (in 1970s dollars, not 2005$) that had to be repudiated only a few years later, because they failed to anticipate how consumers and technology might respond. The real wild card in all this is "peak oil", and if that's what le Caisse D'Epargne is thinking, then Mr. Hubbert has won another convert. Whatever their rationale, you have to salute their courage for straying so far from the herd.