During the energy boom, oilmen used the analogy of an accelerating treadmill to explain why they were powerless to reverse soaring prices – they were getting progressively less for their exploration dollars. Between 2000 and 2008, spending quadrupled but supply growth barely budged. Meanwhile, fresh demand from the developing world gobbled up what little could be eked out. Since about half of the world's oil production comes from slowly-declining giant and supergiant fields and about a third from those discovered prior to the 1970s, more and more small, complex fields would have to be exploited just to maintain output.

This argument suffered two blows recently. One was last year's price collapse, indicating that demand is not quite so inelastic. Less appreciated is a wave of discoveries in areas opened up by technological advances. Salt formations deep off Brazil's coast have the potential to be a new North Sea, while deepwater discoveries have been made in the Gulf of Mexico and offshore Africa. Added to this are shale formations that have turned a US natural gas shortage into a glut. Similar geological formations worldwide might mean that potential hydrocarbon reserves have been underestimated. None of these would have been viable without advances in seismic modelling and drilling technology spurred by high prices.

So, even if the “peak oil” theory were geologically sound, should today's supply forecasts be recalibrated for high prices and human ingenuity? Not so fast, argues energy specialist Simmons & Co. Large offshore discoveries are expensive and have often peaked very quickly, while shale plays may not offset declines in conventional fields. Even if they do, they would require giant infrastructure projects. Replacing existing gathering systems and updating a fifth of refineries might cost $17,000bn.