Friday, December 08, 2006

Curve Ball: Extracting Geological-Economic vs. Geographical-Political Impact on Oil Prices

The Oil Drum had an interesting post today that dredged up some of my old (and I think still accurate) thoughts on the Oil Futures Price Curve. The more I think about it, the more it becomes clear to me that there is very important and very fundamental information available from the shape of this curve, though I don't have the data at this time to prove or disprove my own hypothesis. Three forces impact the curve of oil price futures: Geological-Economic Supply and Demand Balance, Geographical-Political Supply and Demand Balance, and Arbitrage.

Geological-Economic Theory Supply and Demand Balance: Much more predictable out several years, as known projects scheduled to come on line can be forecast several years out, known or predicted geological depletion rates can be applied to existing production, contribution from alternative or non-traditional energy sources can be accounted for at varying price levels, etc. Geology and Economic Theory present a low-noise, digital picture. By that, I mean that day-to-day events in the news have minimal impact on the price as impacted by geology and economics—these forces use broad and general theories to make significant and long-term predictions about what will happen to prices. One example is the “classical” economic prediction that commodity prices will always go down in the long-run. Another example is the “peak oil” hypothesis that supply will begin an inexorable decline, coupled with the economic hypothesis that the elasticity of demand for oil is so low that price will always increase in the face of declining supply. In this sense, the impact of geological-economic factors on prices is digital: either future prices will continually increase or they will continually decrease, depending on which theory one subscribes to.

Geographical-Political Supply and Demand Balance: This is primarily a short-term impact on futures prices due to the imprecise nature of long-term, geopolitical forces acting on supply and demand. Pundits and traders don’t even try to predict the difference between 3-years in the future and 5-years in the future with regards to the impact on crude oil supply due to events in Iran, for example. Predictions are primarily for what will happen in the next year, with impact of future predictions having exponentially less impact to the degree that geopolitical predictions have virtually zero impact on prices more than two years in the future. Geopolitics present a highly “noisy” signal that diminishes sharply as it travels into the future.

Arbitrage: This force pulls the curve towards a mean. Even IF we knew that oil supply would suddenly drop by 75% exactly five years from today, the price of an oil future 5 years out is structurally prevented from vastly outpacing the current price. This is because if prices 5 years out were $30/barrel above prices for next month, arbitrageurs would buy near-term contracts and take delivery while simultaneously selling contracts for 5 years into the future. They would then store the oil until they had to deliver it to cover the future contract, and pocket the difference in the price of those contracts minus the cost to store. The ability to arbitrage is, of course, itself constrained by the availability of storage facilities—but this is only a very near-term constraint because unmet demand for storage facilities will result in more being built.



Reading the Curve: It is my hypothesis that the shape of phase one of the futures price curve is primarily dictated by geopolitics, and that phase two of the curve is primarily dictated by geology & economic theory. If true, then we can extract geopolitical uncertainty by looking at phase one, and we can extract the market consensus on geology and economic theory as it relates to oil production by looking at phase two. This is a critical distinction because it corrects the popular fallacy that current-month price movements are an indication of long-term oil supply (suggesting that only the second phase of the curve should be used for energy policy and economic viability decisions), as well as telling us when the market consensus has finally accepted peak oil (and thereby a contango curve for oil futures prices). More specifically, it is not the empirical level of the second phase of the curve (its dollar value) that matters, but rather its slope: the price of oil two years out compared to the price of oil five years out.

Analysis of the above figure--if you accept my thinking thus far, which remains only a hypothesis--tells us that there is a short-range geopolitical premium in oil prices, and that there is still a general market expectation that, over the long-run, the price of oil will decrease (as shown by the negative slope in phase 2). However, comparison to phase 2 of this same graph from 6 months ago suggests that the market is much less certain in their convictions regarding this point. I think it is also important to point out that the more important contango vs. backwardation determination for oil markets comes exclusively from phase 2, where the valuable information of contango or backwardation conditions can be transmitted, free from the geopolitical noise. Suggesting that oil markets are now in contango because the current price is lower than the price 5 years out is incorrect, in my opinion, because it ignores the upward translation caused almost exclusively by the near-term geopolitical noise.

2 comments:

Anonymous said...

hi Jeff long time reader first time poster. I don't follow what your saying, you say that the price of oil is only high due to geopolitical reasons and once these disappear the price of oil will fall because classical economics says it should?

Im a bit confused because I thought you were a peak oil believer and this doesn't seem to be in line with peakoilevangelicism

cheers

Jason Godesky said...

I doubt Jeff is as interested in "evangelism" than simply understanding what's going on, but you may want to look at his previous post on backwardation and contango as the economic measure of when the market "gets" Peak Oil. Here, Jeff's carrying on that analysis, showing that getting past some short-time fog, the market still hasn't "gotten" Peak Oil, but it's moving in that direction.