Writing The Prize (1991) , and winning a Pulitzer for it, brought Daniel Yergin automatic creds in the energy industry. Through Cambridge Energy Research Associates (CERA), he has consistently maintained a cornucopian viewpoint about the future availability and price of oil, to the point that the energy depletion community has defined a Yergin unit as the $38 per barrel that in 2005, Yergin predicted would be the steady price of oil. Oil reached two Yergins in 2006, spiked to 3.6 Yergins in 2008, and currently Brent crude is trading at 3 Yergins. 

Yergin is in the contradictory position of claiming both that oil prices will stay low and that there will be more extraction of, "proven reserves" through advanced techniques. Proven Reserves means, "Quantity of energy sources estimated with reasonable certainty, from the analysis of geologic and engineering data, to be recoverable from well established or known reservoirs with the existing equipment and under the existing operating conditions."

But those advanced techniques cost more money per barrel, which can only be supported by higher and higher oil prices or government stimulus. If prices had stayed at $38/bbl, as Yergin predicted, "additions and extensions" to extraction would never have been profitable. In fact, collapsing and rebounding demand make it difficult to know which extraction technologies will cost less than the price of oil.

Nevertheless, in his Saturday Essay in the Wall Street Journal, Yergin still wants to assure us that There Will Be Oil and that Peak Oil will always be pushed back as we shift more oil from not recoverable to proven reserves. He admits that geologist M King Hubbert got the date of the US Peak correct, but claims that: 

Hubbert's original projection for U.S. production was bold and, at least superficially, accurate. His modern-day adherents insist that U.S. output has "continued to follow Hubbert's curve with only minor deviations."

But it all comes down to how one defines "minor." Hubbert got the date exactly right, but his projection on supply was far off. He greatly underestimated the amount of oil that would be found — and produced — in the U.S.

By 2010, U.S. oil production was 3½ times higher than Hubbert had estimated: 5.5 million barrels per day versus Hubbert's 1971 estimate of no more than 1.5 million barrels per day. Hardly a "minor deviation."

What Yergin doesn't explain is that Hubbert's 1956 prediction was only for conventional oil extraction in the lower 48 states. Hubbert could not have included the immense amount of oil extracted from Alaska's Prudhoe Bay - discovered in 1968. But Prudhoe reached peak in 1979 and there haven't been any comparable discoveries in the US. Cornucopians talk about ANWR and coastal drilling, but we've reached a level of consumption where new discoveries would have to be more than immense to make a difference.

Hubbert also did not include deep sea oil from the Gulf of Mexico, like that in the BP oil spill. Hubbert did not include synthetic oils from the environmental disaster of Canadian tar sands. Hubbert didn't include oil drilled or fracked from the shale in the Bakken Formation. Hubbert also didn't include biofuels made from crops that require natural gas to make fertilizer and oil to make diesel fuel.

Extracting unconventional oil is far more destructive to the environment than raising a derrick and drilling a pipe. We as a society seem all-too-willing to countenance the loss of fish, birds, bats, bears and people without influence if it will keep the cars running and the lights on, but are we willing to give up our fresh water aquifers, too?

In the comments, Richard Heinberg wrote:

For decades there have been those who warned of oil shortages and high prices, and those who promised there would be plenty of cheap oil for the foreseeable future. For the first three-quarters of the twentieth century, the "cornucopians" proved right. Then, as oil discoveries declined, country after country began to see peaking and declining production. "Peak oil" analysts successfully forecast these production declines nation-by-nation, and during the past tumultuous decade also were generally right about world oil production rates and prices. Meanwhile, the price and production forecasts of cornucopians like Daniel Yergin have diverged further and further from reality. World oil reserves may be large, but the oil industry is replacing cheap, easy-to-get oil with dirty, expensive substitutes. Under these circumstances, reserves are a poor basis for forecasting future production rates. The "peak oil" analysts evidently understand the complexity of the situation, but Yergin--despite his encyclopedic knowledge of oil industry history--seems not to.

And co-originator of the Export Land Model, Jeffrey Brown, commented:

The EIA shows that global annual crude + condensate production (C+C) has been between 73 and 74 mbpd (million barrels per day) since 2005, except for 2009, and BP shows that global annual total petroleum liquids production has been between 81 and 82 mbpd since 2005, except for 2009. In both cases, this was in marked contrast to the rapid increase in production that we saw from 2002 to 2005. Some people might call this "Peak Oil,” and we appear to have hit the plateau in 2005, not some time around mid-century. 

Only if we include biofuels have [we] seen a material increase in global total liquids production.

In the US, there are some good stories about rising Mid-continent production, and US (C+C) production has rebounded from the hurricane related decline that started in 2005, but 2010 production was only very slightly above the pre-hurricane level that we saw in 2004, and monthly US production has been between 5.4 and 5.6 mbpd since the fourth quarter of 2009, versus the 1970 peak of 9.6 mbpd. Incidentally, US net oil imports of crude oil plus products have fallen since 2005, primarily as a result of a large reduction in demand, because of rising oil prices (which Mr. Yergin predicted would not happen), but EIA data show that the US is still reliant on crude oil imports for two out of every three barrels of oil that we process in US refineries. 

However, the real story is Global Net Oil Exports (GNE), which have shown a measurable multimillion barrel per day decline since 2005, and which are measured in terms of total petroleum liquids, with 21 of the top 33 net oil exporters showing lower net oil exports in 2010, versus 2005. An additional metric is Available Net Exports (ANE), which we define as GNE less Chindia's combined net oil imports. ANE have fallen at an average volumetric rate of about one mbpd per year from 2005 to 2010, from about 40 mbpd in 2005 to about 35 mbpd in 2010 (BP + Minor EIA data, total petroleum liquids).

At the current rate of increase in the ratio of Chindia's net imports to GNE, Chindia would consume 100% of GNE in about 20 years. Contrary to Mr. Yergin’s sunny pronouncements, what the data show is that developed countries like the US are being forced to take a declining share of a falling volume of GNE. In fact, our work suggests that the US is well on its way to “freedom” from its reliance on foreign sources of oil, just not in the way that most people hoped.