Goodbye To All That Oil

Until recently, peak-oil analysts got about as much respect from the energy establishment as do perpetual-motion enthusiasts. But now, with oil prices headed for uncharted territory and even Saudi Arabia seemingly unable to boost production to higher levels, the peak oil idea – which says that world oil production will go into irreversible decline sometime in the the next decade or two – is quickly morphing into conventional wisdom.

Fifty years ago, geologist M. King Hubbert showed that the output of an oilfield, or indeed the oil production of an entire country, increases year by year up to the point (a "peak") at which approximately half the oil is exhausted. From there, he said, annual output drops inexorably toward zero.

Hubbert hit the bullseye with his prediction that U.S. production would peak in 1970. And over the past half century, country after country has seen its oil production hit a peak and start dropping. Yet for decades, economists, petroleum executives and government officials refused to follow Hubbert's analysis to its logical conclusion – that in the easily foreseeable future, humanity will pass over a global peak of oil production, where there awaits a very grim, slippery slope.

The Hubbert Curve, designed by geophysicist M. King Hubbert, illustrates that over time, the rate of oil production rises and then falls in a bell-shape pattern.

But gradually, in the past couple of years, the main issue in the oil debate has shifted from whether a world peak will occur to when. And when it comes to peak-oil predictions these days, there is no shortage.

Please place your bets

Colin Campbell of the Association for the Study of Peak Oil and Gas (ASPO) predicts that production will begin its decline between now and 2010.

British Petroleum exploration consultant Francis Harper believes it will happen between 2010 and 2020. Consulting firm PFC Energy puts it at around 2010 to 2015. The publication Petroleum Review predicts that demand will outstrip supply in 2007. Richard Heinberg, author of the 2003 book, The Party's Over: Oil, War, and the Fate of Industrial Societies, expects a peak in 2007 or 2008.

Retired Princeton professor Kenneth Deffeyes, author of the just-published, Beyond Oil: The View from Hubbert's Peak is more pessimistic, and more specific, about when the peak will happen: Thanksgiving Day, 2005. (His tongue appears to be in his cheek regarding the day, but not the year).

If all that is too gloomy for you, energy consultant Michael Lynch maintains that there's no peak in sight for "the next 20 or 30 years." Peter Odell of Erasmus University in the Netherlands has tacked a full 30 years onto Deffeyes' grim prediction, setting a date of Thanksgiving 2035. And Uncle Sam has the cheeriest news of all: a peak year of 2037 forecast by the Department of Energy.

Now how many times has someone told you, "Oh, yeah, all my life they've been saying the oil's about to run out, and it hasn't done it yet"? In fact, the record of oil forecasting has not been an exercise in Chicken-Littlism.

Asking, "When will oil peak and begin its decline?" (not, "When will it run out?"), the prognosticators of the past came up with dates only five to 10 years ahead of many of today's predictions. Roger Bentley of the University of Reading found that in the 1970s – during the last outbreak of peak-oil fever – analysts from "reputable organizations" (including Esso, Shell, the UK Department of Energy, and the U.N., as well as Hubbert himself) were nearly unanimous in predicting a world oil peak somewhere around the year 2000.

Does the peak year even matter?

With oil prices soaring, economic logic says the sooner the peak's date can be nailed down, the better. Financial web sites are buzzing about it, but in a somewhat merrier key than the peak-oil sites. One research firm is even forecasting production peaks for individual oil companies, with obvious implications for stock values.