December 8, 2004
A year ago, most oil experts thought oil prices would average less than $30 per barrel in 2004. In fact, OPEC actually curtailed production early in the year to absorb what they thought would be an oil surplus.
No surplus ever materialized.
Of course, the supply disruptions in
Iraq, Nigeria, Norway, and even the hurricane impacted Gulf of Mexico caused
much of the miscalculation.
Another error was in the assumptions of
world demand for oil. China
alone raised its oil consumption by a million barrels a day.
U.S. economic growth at nearly 4.4 percent required another half
million barrels. When other
Asian needs, South American growth, and even the mild increases in Europe
are included, oil consumption jumped nearly 3 million barrels a day while
supply disruptions removed 3 million barrels per day from availability.
What has not been fully understood,
however, is the shift in inventories needed to supply the world with
petroleum. The tanker runs are
several days longer to Asia than the rest of the world, Indonesia’s excess
petroleum supply notwithstanding.
Nevertheless, through much of the year,
the experts continued wondering why so much oil was being consumed.
China’s industrial mix is petroleum intensive, but that is not true
of much of the world. Computer programming, electronic games, and other electronics
really do not consume the same amount of energy per dollar of value as China
needs for its development.
The latest oil inventory report suggests
another reason for the miscalculation.
Consumers have been building their own inventory of oil.
They do not have storage tanks in their
backyard. Instead, they have
been filling their tank whenever they see “good” gasoline prices.
In the Northeast, they apparently also were buying heating fuel
Just imagine if all the 220 million
registered vehicles in the United States filled up at half empty rather than
nearly empty. The result would
be every vehicle, on average, carrying another 4 gallons in their car.
Assume further that the vehicle fleet
has shifted to SUVs, with tanks holding 4 more gallons on average than the
passenger car. Of course,
homeowners who rush to fill their heating fuel tanks at the first blush of
cold air could have the same impact upon heating fuel.
Of course, you should ask why people
would want more inventory of gasoline and heating fuel when prices are
rising. Doesn’t this violate
the most basic assumption in price theory: rising prices curtail quantities
demanded in almost all conceivable market conditions?
I once pondered this question for
sometime before coming to the appropriate answer.
If rising prices increase price expectations faster than prevailing
prices are changing, then the current price is actually falling relative to
the unkown, but expected, future price.
Compared to all other goods, people would want less oil.
However, compared to oil tomorrow (the fastest rising price), they
would want more oil today.
By the way, this behavior can be
observed for much more than oil. We
could talk about beach houses or mountain retreats and come to similar
results. Indeed, stock prices and prices of other assets also behave
in this manner.
What this means is that markets with
inventories tend to overshoot their equilibrium values.
Indeed, if many markets are overshooting at the same time, conditions
for curtailing production leading to recession develop.
However, the oil market has a special wrinkle. Producers will store their inventory in the ground (not produce at capacity) if they believe future prices are rising faster than current prices.
I had a student who demonstrated that
these production choices depended upon what producers could earn on
investments relative to what they could gain by not producing today.
In most cases, producers will be slow to raise production if rising
prices lead to higher expected prices.
In short, they restrict supply just as consumers, using the same
feelings, are expanding purchases for storage.