Streetwise Professor

May 8, 2009

What’s Up With the Oil Market?

Filed under: Derivatives,Economics,Energy,Financial crisis — The Professor @ 9:22 pm

Other than the price, that is.  The excellent Izabella Kaminska continues her fascinating coverage of the relation between oil prices and fundamentals at FTAlphaville. Izabella provides additional data on the continuing accumulation of oil and product inventories in the face of rising prices, and links to a Goldman Sachs analysis thereof.  

A positive co-movement between inventories and prices is an anomaly in the standard theory of storage model.  In that model, an negative demand shock (or a positive supply shock) leads to (a) a decline in price, and (b) an increase in inventories.  

I have devised a model (the ideas for which I first explored on SWP) in which you can get a positive covariation between prices and inventories due to stochastic fundamental volatility.  The basic idea is that when fundamental uncertainty goes up, people increase precautionary inventory holdings.  The only way to do this is to bid up price in order to reduce consumption and increase production.  

Although I think that this model can explain some market behavior (such as in the post-hurricane period 2005-2006), it doesn’t seem to fit the current episode.  If anything, fundamental uncertainty is declining of late.  Stock prices have been rallying.  The volatility indices (e.g., the VIX) have been declining.  

Here’s an alternative explanation.  Traditional models posit Markovian processes for (net) demand shocks.  This means that current demand shocks imply a distribution of future demand.  Moreover, the models are typically stationary, meaning that a given current demand shock leads to a smaller change (in the same direction) in expected future demand.  In these models, the distribution of future demand changes only if current demand changes.  

An alternative model (which is hard to implement formally) incorporates “future” shocks that are distinct from current shocks.  (A Fama-French JF paper on metals prices from 1988–if I recall the year–explores this idea.)  That is, information may arrive today that has no implications for current demand or supply, but does has implications for future demand or supply.  So, for example, if new information arrives that suggests the economy will be better in 6 months than than had been previously anticipated, a rational economic response is to add more to inventory in anticipation of higher future demand.  Expected future prices rise due to the bullish future information, and in conditions of full inventory, the cost-of-carry relationship causes current prices to rise too.  (Indeed, holding current demand constant, the higher price induces lower consumption and higher production, which allows the accumulation of additional inventory.)  

This is pretty close to the Goldman-Sachs story.  Goldman also argues that if oil storage facilities fill up, the connection between future prices and spot prices will break down.  In the context of the model, if the marginal cost of storage becomes very steep, the increase in future demand leads to a rise in the future price, but a much smaller rise in the spot price; it is too costly to accommodate the future demand shock by adding even more to inventory due to the very high and rapidly rising cost of storage.  Thus, current inventory (and hence current consumption) do not respond to the future demand shock; since consumption doesn’t respond, spot prices don’t change either.  

Note that the nature of the “future shock” (to steal the name of an old book) is not specified here.  It could be a real shock or a nominal one.  That is, it could be that information arrives that suggests that real demand will increase in the future, or that only nominal (money) demand will increase.  That is, inflation is expected to increase.  This could lead to increases in nominal spot and futures prices, even in the face of stagnant or declining real demand.  

One piece of evidence that could bolster this interpretation is that nominal Treasury yields are beginning to rise.  The 30 year auction yesterday was something of a disaster, and 10 and 30 year yields are rising. Indeed, over the past week when oil prices have been rising, the breakeven rate (the difference between 10 year nominal T-note rates and the corresponding TIPS yield) has increased from about 1.41 percent to 1.58 percent.  Not a big deal, but the direction is interesting.  

So, the oil market could be a “green shoot” signaling an impending recovery of demand.  One not quite here yet, but on the horizon.  Or, it could be a “green” shoot in the sense that (a) money is green, and (b) per Milton Friedman, inflation is always and everywhere a monetary phenomenon.

Print Friendly, PDF & Email


  1. Couldn’t it also be a miscalculation? I see no reason to expect demand or inflation to be drivers of the oil price over the next couple of years, I expect that deflation will be the dominant theme. I seem to be in the minority, but it would be equally plausible (given recent market miscalculations, think 2007 equities), that market players (and many economists) have far too much faith in the ability of central banks to inflate in a hyper-deflationary environment (in an Austrian money supply sense, meaning, I believe that if the money supply were properly measured, it would be contracting on a massive scale, if credit were being marked to market).

    I believe that market participants are hoarding supplies, hoping that prices will rebound above average marginal costs (which I believe is around $70). They can only do this for so long, and ironically, their hesitance to bring supply to market will increase the deflationary spiral, as they will likely all be selling into the market at the same time.

    Is this story unreasonable?

    Comment by Danny — May 8, 2009 @ 9:57 pm

  2. Thanks for expanding on my post, I very much like the analysis. Regarding your concluding points, I have to say I am of the opinion it is down to the money is green factor. What I hear from the market is that most of buying at the front end has been down to funds. I urge you to have a read of this post too:

    Comment by Izabella Kaminska — May 9, 2009 @ 4:47 am

  3. Danny, great observation.

    The Canadian oil sands need $60 per barrel for their extraction and processing costs to make a profit. That’s the cost of the most expensive oil to recover. Extrapolate from there based with the world supply of cheaply recoverable/processed/refined Sweet Texas Crude and OPEC below that.

    I think inflation versus deflation is what’s being bet upon. And, never discount the stupidity of the Dems/Obama to do the wrong thing in punishing Big Oil profits and drive down that sectors share prices in the near term.

    I’ll be the first to admit I’m not an economist.

    Comment by penny — May 9, 2009 @ 7:20 pm

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress