have to go right to achieve the IEA's and EIA's benign vision of 2030—a balancing
of world oil supply and demand with real oil prices only modestly
higher. I cannot forget how wildly wrong the U.S. Department of Energy
was in its 1979 forecast of $150-a-barrel oil in 1995 (in 2006 prices).*
*The DOE's basic miss was to underestimate the long-term price elasticity for oil. It is price
elasticity that determines the price change needed to converge supply and demand. Obviously,
459
THE AGE OF TURBULENCE
To achieve the twin goals of enhanced national security and curtailed
global warming, the growth rate of U.S. petroleum consumption must flatten,
and eventually consumption must decline outright. The big opportunity
for displacement is on America's highways, where one out of every seven
barrels of petroleum consumed worldwide is burned: 9.5 million barrels per
day in gasoline and 2.5 million barrels per day in highway diesel in 2005.The
latter is consumed by the nation's eight million heavy trucks, which average
less than seven miles to the gallon. By themselves, those heavy trucks consume
as much petroleum as all of Germany. Only China and Japan, and of
course the United States, consume substantially more.
In looking to the future, we are not likely to find the answer to the
question "What does OPEC need to do?" as useful as the answer to the
question "What are they likely to do?" Attempts to answer "What needs to
be done?" have a decidedly mixed historical record of accomplishment. I
would thus be more inclined to accept the IEA "contingency forecast,"
which assumes that OPEC will lag in expanding its crude-oil capacity. The
consequence, according to the IEA, will be an average world price of $130
per barrel ($74 in 2005 prices) versus about $50 in 2005. Oil demand in
2030 in this scenario remains surprisingly strong—it is still 109 million barrels
per day, up from 84 million in 2005. (The IEA "reference case," in
which OPEC is assumed not to lag, foresees 116 million barrels per day.)
This is not a shock scenario, and there are lots of scenarios that anticipate far
worse.
I trust that at the end of the day we will allow markets to guide our
preferences in reducing petroleum consumption. Our experience with oil
rationing, as I noted earlier, has been poor.* Another way to curb consumption
would be a gasoline tax of, say, $3 or more per gallon, phased in over
five or ten years with the resulting revenue used to lower income or other
taxes. I come very reluctantly to taxes as an alternative way to accomplish
what competitive markets could do. But while oil markets are highly competitive
in the developed world, the market approach is clearly vulnerable
the lesser the price elasticity, the more prices have to change to balance projected supplies and
demands.
*Rationing did seem to work in the United States during World War II, but even then, black
markets were extensive.
460
THE LONG-TERM E N E RG Y S O U E E Z E
in a world where a single act of terrorism can shut down massive chunks of
oil production and cripple the global economy. There is no insurance, or
hedging strategy, that can defend against that. We often forget that to function
effectively, a competitive market must be voluntary and free of significant
threats of violence, and that trade must be unencumbered. Remember,
markets are not ends in themselves. They are constructs to assist populations
in achieving the optimum allocation of resources.
We need significantly higher gasoline prices to wean us off gasoline-
powered motor vehicles. The geopolitical price premium is apparently not
large enough to do that unassisted. The expectation of higher gasoline
prices through taxes (or an oil-supply squeeze) would galvanize large technical
breakthroughs in the production of ethanol. Corn ethanol, though valuable,
can play only a limited role, because its ability to displace gasoline is
modest at best. One bushel of corn yields only 7.2 gallons of ethanol, which
means that all 11 billion bushels of corn that the United States produced
in 2006 would have yielded only 5.2 million barrels of ethanol a day, the
energy equivalent of 3.9 million barrels a day of gasoline, or only a third
of U.S. highway use, and less than a fifth of the 21 million barrels a day
Americans consumed in 2006. And, of course, if all corn were devoted to
ethanol, our pigs would starve. Cellulosic ethanol derived from switchgrass
or agricultural waste seems to hold greater promise. A joint study by the
U.S. Department of Agriculture and Department of Energy credibly estimates
that fuels derived from plant matter, or biomass, have "the potential
of sustainably supplying much more than one-third of the nation's current
petroleum consumption." Other countries' use of biodiesel fuel, which is
derived from vegetable oils and other sources, can add to the displacement
of OPEC petroleum.
Alternatively, if ethanol fails and if gasoline prices are high enough,
plug-in hybrids will significantly displace petroleum consumption over
time. Battery technology is making gradual progress; there is already ample
electric power being generated to supply plug-ins, particularly if power
companies move more toward peak-load pricing. If we can shed our fear of
nuclear power, the concern that plug-in hybrids will ultimately be powered
from conventional electric utilities burning polluting coal will be resolved.
Conventional hybrid cars, cars running on cellulosic ethanol, and plug
461
THE AGE OF TURBULENCE
ins could displace a major part of the petroleum burned on U.S. highways.
Wider use of more efficient diesel engines could induce further significant
displacement of petroleum. But to speed such displacement would require
either a vast increase in supplies of cellulosic ethanol or very expensive
gasoline. Taxes can ensure the latter. I consider the argument that gasoline
tax hikes are politically infeasible irrelevant. Sometimes the duty of political
leadership is to convince constituencies that they are just plain wrong.
Leaders who do not do that are followers.
A gasoline tax would not impose a very large burden, especially if
phased in over a number of years. U.S. household motor fuel outlays, at
3 percent of disposable income in early 2007, are where they were from
1953 to 1973 and far below the 4.5 percent experienced during the crisis
of 1980. Even at heights of $3-plus per gallon in July 2006, motor fuel
consumed only 3.8 percent of disposable personal income. Yet Americans
are very sensitive to gasoline prices. We complain when they rise. Americans
nonetheless continue to drive as much as before. In the face of gasoline
price spikes, they reduce mileage driven only for a short while. The
average number of miles driven per licensed driver has continued to drift
upward: from 10,500 miles per driver in 1980 to 14,800 miles in 2006, an
increase of 1.3 percent per year. With higher prices, since 2002 the increase
has slowed to 0.2 percent per year. Drivers consume less gasoline only because
they eventually buy more fuel-efficient cars.
It should be obvious that as long as the United States is beholden to
potentially unfriendly sources of oil and gas, we are vulnerable to economic
crises over which we have little control. Petroleum is so embedded in today's
economic world that an abrupt severance of supply could disrupt our economy
and those of other countries. U.S. national security will eventually require
that we see petroleum as an energy source of choice, not necessity.
The burgeoning global economy devours vast amounts of energy. Despite
the dramatic fall in the amount of oil, and more generally energy,
consumed per dollar of world output, all credible longer-term forecasts
conclude that to continue on the path of world growth over the next quarter
century at rates commensurate with those of the past quarter century
will require between one-fourth and two-fifths more oil than we use today.
462
THE LONG-TERM E N E RG Y S O U E E Z E
Most of this oil will have to come from politically volatile regions because,
as we have seen, that is where most of the readily extractable oil resides.
What do governments whose economies and citizens have become
heavily dependent on imports of oil do when the flow becomes unreliable?
The intense attention of the developed world to Middle Eastern political
affairs has always been critically tied to oil security The reaction to, and reversal
of, Mossadeq's nationalization of Anglo-Iranian Oil in 1951 and the
aborted effort of Britain and France to reverse Nasser's takeover of the key
Suez Canal link for oil flows to Europe in 1956 are but two prominent historical
examples. And whatever their publicized angst over Saddam Hussein's
"weapons of mass destruction," American and British authorities were
also concerned about violence in an area that harbors a resource indispensable
for the functioning of the world economy.
I am saddened that it is politically inconvenient to acknowledge what
everyone knows: the Iraq war is largely about oil. Thus, projections of world
oil supply and demand that do not note the highly precarious environment
of the Middle East are avoiding the eight-hundred-pound gorilla that could
bring world economic growth to a halt. I do not pretend to know how or
whether the turmoil in the Middle East will be resolved. I do know that the
future of the Middle East is a most important consideration in any long-
term energy forecast. Even though oil-use intensity has been significantly
reduced, the role of oil is still such that an oil crisis can wreak heavy damage
on the world economy. Until industrial economies disengage themselves
from, as President George W. Bush put it, "our addiction to oil," the
stability of the industrial economies and hence the global economy will remain
at risk.
463
TWENTY-FIVE
THE DELPHIC FUTURE
P
P
eople have always been enthralled by the notion that it is possible to
peer into the future. Ancient Greek generals sought audiences with
the oracle at Delphi to guide their military adventures. Fortune
tellers thrive to this day. Modern Wall Street employs phalanxes of very
smart people to read what the entrails of market performance say about
future stock prices.
To what extent can we anticipate what lies ahead? Inbred within all of
us is the capacity to weigh probabilities, a gift that helps to guide our actions
in everything from the mundane to matters of life and death. These judg
ments are not always right, but they have manifestly been good enough to
enable humans to survive and multiply. Modern economic policymakers for
malize such decision making in mathematical terms, but humans were judg
ing probabilities long before we invented the math to explain them.
Fortunately for policymakers, there is a degree of historical continuity
in the way democratic societies and market economies function. This en
ables us to reach back into the past to infer inherently persistent stabilities
that, while not having the certainty we attach to physical laws, nonetheless
TH E DE LPH IC FUTU RE
offer a window on the future that is more certain than the random outcome
of a coin toss. There is indeed much that we can infer about the U.S.
economy and the world at large in the decades ahead, especially if we
adopt Winston Churchill's insight: "The further backward you look, the
further forward you can see."
Most legal and economic institutions change slowly enough to facilitate
anticipating future outcomes with some reasonable degree of probability.
Nonetheless, a large body of academic literature exists that questions
how successfully people can forecast financial outcomes. Proponents of "efficient-
market theory" have famously argued that all publicly available information
that would induce a stock-price change is efficiently factored by
the market into the current price of the stock. Hence, unless an investor has
special or inside knowledge not available to the market at large, he or she
cannot anticipate price changes. As evidence, they point to the well-known
inability of managed equity mutual funds to outperform the S&P 500 consistently.
The evidence that some investors do consistently beat the market
year after year is not surprising. It's just what one would expect. Even if investment
results are purely a matter of chance, a small number of investors
will do exceptionally well—as well as the lucky coin tosser who turns up
heads ten straight times. The probability of ten consecutive heads is 0.1
percent; thus, when you have millions of coin tossers, or investors, in the
end there will be thousands of very successful practitioners of coin tossing,
or stock picking.
Yet the theory of efficient markets cannot explain stock-market crashes.
How does one make sense of the unprecedented drop (involving the loss
of more than a fifth of the total value of the Dow Jones Industrial Average)
on October 19, 1987? As a newly anointed Fed chairman, I was watching
the markets very closely. What new piece of information surfaced
between the market's close at the end of the previous trading day and its
close on October 19? I am aware of none. As prices careened downward
all that day, human nature, in the form of unreasoning fear, took hold, and
investors sought relief from pain by unloading their positions regardless
of whether it made financial sense. No financial information was driving
those prices. The fear of continued loss of wealth had simply become
465
THE AGE OF TURBULENCE
unbearable.* And while the economy and corporate profits subsequently
advanced, it took nearly two years for the Dow to recover fully.
When markets are behaving rationally, as they do almost all the time,
they appear to engage in a "random walk": the past gives no better indication
