Dear Mr. Porter:

I started reading your paper thinking that here we would have an authoritative statement of this important subject. I began to jot down notes and comments on points of detail, but by the time I was about half way through, the penny dropped. I realized that it was not an objective and analytical study as it appeared to be, but a "lobbying" document to deliver a predetermined message and discredit any counter arguments. In some ways, it reminds me of a trial in which counsel tries to undermine the evidence by subtle innuendo and inference. I will comment only on the main flaws in the logic.

2. Confusing the USGS Study
The USGS study is itself very sound once its special reserve definitions are decoded. A simple way to do that is to explain that Proved Reserves have a high probability of occurrence (90-95%) whereas that the USGS "Identified" reserves have a low probability (5-10%), being notional geological amounts unconstrained by economic, technological or timing factors. "Reserve Growth" is treated as if it were a dynamic akin to discovery, when it is mainly an artifact of the reporting process. Using Median Probability reserves, in which the risks that the actual will be above or below the estimate are equally matched, overcomes all these confusions. If the current discovery trend (about 7 billion/year) is taken into account, it is clear that only the low end of the USGS Yet-to-Find deserves consideration to the extent that it could affect the pending production peak. The USGS number of 2.3 trillion barrels reduces to about 1.75 once these factors are taken into account. This is a very good number based on distribution and discovery trend analysis.

3. Discrediting Hubbert
Having admitted that Hubbert hit it on the nail when he predicted that the USA would peak when it did, efforts are made to discredit his approach by magnifying a segment of his decline curve and emphasising how the actual curve has differed. Hubbert's contribution was to equate peak unfettered production with the midpoint of depletion. That there are relatively minor daily departures from the predicted curve, 40 years on, hardly diminishes the value of his general proposition.

4. The Reserve to Production Ratio escape hatch
Having admitted that Hubbert got it right, and having accepted that about two trillion is a good consensus Ultimate Recovery, logic would lead to the conclusion that peak production will come when one trillion barrels have been produced. Instead, appeal is made to the mistaken notion that dividing the reserves by current annual production indicates the number of years current production can be sustained. Since Production and Reserves fall in parallel the R/P ratio stays the same. It is an utterly flawed approach, widely used in the report. After peak, production declines at a certain depletion rate until it is exhausted, as every oilfield demonstrates.

5. Failure to distinguish conventional and non-conventional oil
More than 95% of all oil produced to-date, as well as that to be produced over the next decade or so, is what can be called conventional. Once it has peaked and shortages appear, more non-conventional oil will be produced by tapping large heavy oil deposits, and undertaking more enhanced recovery techniques, infill drilling, etc. But this kind of oil has a very different depletion pattern, rising only slowly to a long low plateau, and is costly to produce, also in environmental terms. It may amount to about half of all production by 2050, but by then the total will be less than half what it is today if the two trillion consensus number is about correct. We are of course not running out of oil, the question posed in the title: only cheap and abundant supply.