In 1956 M. King Hubbert predicted that oil production in the US states (the 48 lower US states to be precise) would peak in the early 1970s. His pronouncement was greeted with derision but proved quite accurate. He also predicted that world oil production would peak in the 1990s – not knowing then that the oil crisis of 1973 was coming. &ldquoHubbert's peak” is the point at which half the recoverable oil has been produced. After this point he predicts that oil production will decline in a mirror image of the increase leading to the peak. Some industry experts believe we have indeed now reached peak oil production. After just 150 years or so we are half way through our recoverable stocks of oil but oil consumption is expected to keep increasing.

But haven't we been here before? Were we not warned years ago that we were on the brink of running out of oil? And yet the world's supply did not dry up. Not everyone agrees that the future looks grim, as indicated by the titles of some essays on the subject: &ldquoClosed Coffin: Ending the Debate on 'The End of Cheap Oil',” and &ldquoCrying Wolf: Warnings About Oil Supply,” (Michael Lynch); &ldquoFixed View of Resource Limits Creates Undue Pessimism” (Michael Lynch and M.A. Adelman); and &ldquoWe will never run out of oil” (Michael Moffat).

So how much oil is there? First of all, it is important to make a distinction between conventional and non-conventional oil. The former is what we have been using up till now: oil that is relatively easy to extract. Non-conventional oil means oil that is difficult and expensive to exploit, for example: heavy oil, tar oil, oil found in hostile environments and oil located in deposits that are too small to be economically viable. The border line between the two types of oil is somewhat fuzzy but important to bear in mind as it may go some way to explain the widely varying estimates of oil left in the world. Peter Odell has estimated the reserves of non-conventional oil at between 2 and 5 trillion barrels, the reason for the wide range being that non-conventional reserves are much less thoroughly examined than conventional (1983, 109).

Economists repeatedly criticise Colin Campbell of the Association for the Study of Peak Oil and Gas for underestimating the amount of oil there is and pushing the year of peak oil production ever further back. When I asked him about this over the phone he admitted that his earlier studies produced inaccurate results but explained: &ldquoIt's that the public data on reserves is grossly unreliable. The first of these studies that I published initially was done for the Norwegian government on the basis of public reserve numbers and I have to confess that I myself did not realise the degree to which companies underreported what they found. … It was only later that I got access to the industry database which listed everything by field.”

The 2004 BP Statistical Review estimates global reserves at 1,147 billion barrels and according to IHS Energy Group's World Petroleum Trends Report 2003 an average of 7.42 billion barrels is being found per year (Salameh 2004, 27-28). The world consumes roughly 29 billion barrels a year; this is expected to rise to 51 billion barrels by 2035 (Roberts 2004, 7).

However, nothing is as simple as it seems in the matter of oil production and depletion. For example, Canada reclassified non-conventional oil produced from tar sands as conventional oil in 2002 causing a jump in its reserve estimate though no new oil had been discovered (Hallock et al 2004, 1677). Also, question marks hang over the reserves claimed by the OPEC countries. Iraq reported reserves of 100 billion barrels every year from 1985 to 1998 despite drilling no new wells while all the time producing oil (Deffeyes 2001, 176). When OPEC decided to base its quotas on oil reserves instead of production a number of member states produced totally new reserve estimates (Salameh 2004, 26). Nor is this chicanery confined to the middle East: Shell has admitted overstating its reserves by 20%. When oil people talk about reserves they generally mean recoverable

reserves (“URR” stands for Ultimate Recoverable Resource) and on this

distinction hinges an important difference of opinion between – broadly

speaking – geologists and economists.

For a branch of study sometimes known as &ldquothe dismal science,” economists such as Adelman, Lynch and Odell are remarkably cheerful about the future of oil. Odell even discusses the theory that oil is a renewable resource (Odell 2000, 199). When I asked Michael Lynch about this he wrote: &ldquoOil is effectively a non-renewable resource. (About 3 million barrels a year are created by geological processes, according to current estimates, which is trivial.) But oil reserves are a renewable resource, as they are replenished by drilling. M.A. Adelman often makes this point, but it confuses some not familiar with the nomenclature.”

The confusion is forgivable: Adelman writes in The Genie out of the Bottle, &ldquoOil is a renewable resource, with no intrinsic value over and above its marginal cost, of which a small fraction is current operating expense” (34). Presumably this is intentionally provocative and it leads to the

fundamental difference between the geologists and the economists in this

debate: “URR” means not how much oil there is, but how much oil we can get

at a price worth paying. Geologists, notably Colin Campbell and Kenneth

Deffeyes, take what would seem to be the common sense view that there is a fixed amount of oil in the earth, which will not be renewing itself any time soon. Furthermore, we have to discover it before we can produce it: the production cycle follows the discovery cycle with a time-lag of several years. At a conference on &ldquoIreland's Energy Security of Supply” in Dublin in February Campbell showed evidence of this pattern in Russia, Indonesia, Egypt and the UK.