A few years ago, researching a book on cars, it struck me that, although nobody ever mentioned it, oil was a finite resource. How much, I wondered, was left?
Petroleum geologists agree on the answer, given our present consumption was almost unbelievable: the Big Rollover, the jargon term for when the oil production peaks, between now and 2020. After that, although as much will be left as we have already used, it will become steadily more difficult and expensive to extract. And consumption is rising by per cent a year. Soon, demand will outstrip supply.
What will that will do to prices? It will affect every aspect of our lives, where we live, how we shop, how our food is grown, All those depend upon cheap oil. Why weren't people writing about this? Why wasn't it front-page news?
Well, here is the answer; and it is sobering. Geologists have been talking about The Big Rollover since M King Hubbert forecast in 1956 that domestic oil production in America would peak between 1968 and 1972; the actual date was 1970.
But the power rests with the politicians, and they know, better than anyone, there aren't many votes in gloom. Enforce conservation? Raise petrol taxes? No, pols listen to economists, who tell them that as soon as anything gets expensive enough, previously disregarded alternatives appear. But oil is different. Yes, there are alternatives; yes, they are being developed, though not with anything approaching the necessary urgency. But they won't be as cheap as oil, because the physics are against it.
The key to understanding this is the energy/profit ratio; energy returned on energy invested. The benchmark is the oil that bubbled from wellheads in the 1940s discovery boom: drill and out it gushed.
If the energy profit ratio on that was 100 (that is, 100 units of energy gained for every unit invested), by the 1970s it was down to 8 on new discoveries, though production from existing wells was 23 in profit. The profit ratio was 80 on coal at the minehead in the 1950s, 30 in the 1970s; on oil shale, one of the touted future possibilities, it varies from 0.7 (an energy sink: more energy must be invested than will be recovered) to 13.3.
These are fossil fuels. After use, they are gone forever, fatally warming the atmosphere en route. But as we know, this is not the way forward. That lies in renewables: wind, solar, biomass. And here the news is less good. Cars can be run on alcohol, ethanol and methanol. But the energy ratio of ethanol made from sugarcane is at best 1.7. The figure for methanol, made from wood, is 2.6, slightly better, but not much.
And what of hydrogen, fuel of the future? It is not an energy source, but a concentrated way of storing and transporting the electricity produced by other means, wind, or solar cells. It can be used in car engines, and is superabundant. It cannot be mined or pumped but made.
The fuel most often used in this process now is natural gas, among the most energy-inefficient ways possible to use this natural resource.
Bear all this in mind next time a politician tries to tell you that wars in the Middle East oilfields are nothing to do with oil.
Richard Heinberg: The Party's Over - Oil, War and the Fate of Industrial Societies, Clairview, £11.95