Natural Gas prices have been all over the place for the past couple of years, ranging from just over $2 per million British Thermal Units (MMBtus) to over $13/MMBtu. Peter Hall effectively uses supply and demand to explain the wild swings:
So far, a classic case of elementary economics: tight supplies boost prices, high prices boost investment, increased production brings prices back to equilibrium. But are today’s sagging prices actually in equilibrium? Not exactly. Unfortunately for the industry, increased supply coincided with the largest economic contraction in six decades. As such, demand for gas has swooned by 5% to date this year, on a 13% drop in industrial demand. For the moment, prices are abnormally low.
Prices are expected to remain subdued while the world economy works through the current period of weakness. In response to soaring inventories, gas producers have scaled back activity, but are expected to ramp up again as the recovery takes hold. As we march through 2010, prices are expected to climb back to the US $5-6/MMBtu level, enough to kick-start production again.
The bottom line? Surging supply and stifled demand have pummelled natural gas prices. Consumers will enjoy the brief hiatus, but producers can look forward to more normal pricing in 2010.
In the above explanation, Peter Hall is assuming that decision-makers have somewhat myopic expectations (or what are typically referred to as adaptive expectations). In the first paragraph, he assumes that investors expect prices to remain high and hence overinvest; there is no anticipation on the part of the investors that other people will be expanding capacity as well and that prices will fall. I.e., there is a systematic error in the expectations. Rational expectations are out the window.
And then, again, when the price is low, investor-producers do not invest on the basis that they expect the price to rise to $5-$6/MMbtu. Instead, they wait to see if the price actually does rise that far.
These cycles remind me of the old myopic-theory of agricultural prices for the corn-hog cycle and that was (and perhaps still is) taught as the cobweb theory. [note: this Wikipaedia explanation is very good]
I tend to like adaptive expectations models, given my own inability to formulate rational expectations in many instances. And this is probably a fairly good example of that application.
Addendum: James Hamilton has an equation here that captures this adaptive expectations model with a simple lag structure. For those who are interested, note the alternating signs on the lagged variables, which seem consistent with a cobweb model.
Update: it is really clear that these clowns do not understand the cobweb (adaptive expectations) model. Contrary to their assertion, it involves continued overshooting.
Comments