Skip to content

‘Catch-22’ — the paradox of peak oil in a volatile market

Crikey, 13 January 2015

The spectacular oil-price crash following the Saudi-led decision of the OPEC cartel to maintain production — despite a glut and in the face of softening demand — has left peak-oil theorists in an awkward position. Wasn't the cheap oil meant to be running out?

For many market-watchers, the rise of unconventional oil and gas in recent
years — spurred by higher prices and improved technology, especially
directional drilling and fracking — has rendered the so-called "peakists"
irrelevant. In an influential 2012 piece, Business Spectator's Alan
Kohler called the "death
of peak oil
", sparking a terrific
with one of Australia's foremost peakists, Matt Mushalik.

A retired civil engineer who is a member of the Association for the Study of
Peak Oil and who blogs at, Mushalik is a tireless
correspondent to journalists and politicians, critiquing energy and
infrastructure policy from his Sydney home. He also knows his stuff, compiling
the fascinating chart below, which shows incremental crude oil production (that
is, additional supply above a base level) country-by-country, based on official
US data:

Mushalik explains: "When you look at the statistics, outside North America —
excluding US and excluding Canada — crude oil production is basically flat, on
a bumpy plateau, that's a fact."

"The way I see the peak oil story is that conventional oil peaked around
2006," said Mushalik, citing as evidence similar comments made by International
Energy Agency chief economist Fatih Birol to the ABC's Catalyst program in

"The response of the system — I mean governments, banks and the oil industry
— was twofold: firstly, quantitative easing started, [which] enabled the
economy to pay for higher oil prices, and secondly, a certain percentage of
quantitative easing money went into the oil sector as cheap money, hence the
[rapid development of] shale oil and tar sands," Mushalik told

"After the GFC, oil prices went up again to US$110 for three years and that
killed demand. That is why I'm saying the experiment of offsetting the
conventional oil peak with expensive shale oil and tar sands failed."

Like most
analysts, including Birol at the IEA
, Mushalik expects lower prices will
quickly kill off investment in higher-cost, unconventional oil production around
the world, especially US shale oil, which is widely regarded as the Saudis’ main
objective. Supply will fall, so oil prices will go up again. How quickly is
anyone's guess and depends on whether the price fall drives economic growth, in
turn pushing oil demand higher.

"I don't think that anyone has a model to know which time lag there is
between oil price and demand going up because there may be other reasons why
demand is down, not just because of the oil price," said Mushalik, pointing, for
example, to economic slowdown in Europe and China.

Mushalik describes it as a “catch-22”: the world economy chokes on
high-priced unconventional oil, but when oil prices fall, and the economy starts
to pick up again, “then again the oil price is too low to finance all the
investment to stop geological decline," which refers to a decline in the amount
of oil left in the ground.

"So yes, some people say 'oh, the oil price will find its own level between
these two extremes — very high oil prices and very low oil prices', but I don't
think that anyone has a model." So long as economic growth is dependent on
fossil fuels, we will suffer the gyrations of an increasingly volatile oil
market, with ramifications for both private investors and governments trying to
balance their budgets.

Part of the problem is what oil industry whistleblower Chris Cook, a former
director of the International Petroleum Exchange and research fellow of the
University College, London, calls the financialisation of the oil markets, in
which prices are no longer driven simply by supply and demand fundamentals and
are not necessarily a reflection of the scarcity value of oil. Cook claims to be
one of the very few who correctly predicted the oil price crash, as he wrote in
the Asia
just before Christmas:

"I was honored in November 2011 to be a plenary speaker at a conference
convened in Tehran by the Institute for International Energy Studies. I outlined
— to a politely disbelieving audience — precisely how the global oil market
was being financially manipulated and by whom. Moreover, I forecast then, and
many times subsequently, that the oil market price would inevitably collapse to,
and probably temporarily through, a price level of $60 per

this 2012 piece
Cook tipped the oil price would crash to US$75 a barrel as
the Federal Reserve unwound its money-printing program, called quantitative
easing (QE) — it took a little longer to unwind than he expected.

The biggest driver? In what may be a conspiracy theory, Cook argues the
Saudis have worked with investment banks to monetise their reserves early,
selling large quantities of oil using "prepay" contracts — not equity, not
debt, not a forward sale or futures contract, but more like a gift card — to
risk-averse managed funds (he calls them "muppet" funds) looking for a hedge
against inflation (see this piece, which
has an interesting comment thread). Cook says the all-powerful Saudi oil
minister Ali al-Naimi let the cat out of the bag in an interview before
Christmas with the Middle-East Economic Survey (MEES), which was reported around
the world, particularly that the Saudis would not cut production even if prices
hit US$20 a barrel — even if it meant the kingdom was losing money at that

Al-Naimi told
the interviewer
: “A deficit will occur. But what resources do you have in
the country? We have no debt. We can go to the banks. They are full. We can go
and borrow money, and keep our reserves. Or we can use some of our

The upshot? Cook argues there is a strong correlation between QE and oil
prices, charted below, and there is now a window of opportunity to restore some
fundamentals to oil markets:

European Tribune

Whether you buy Cook’s theory or not, the intertwining of the financial and
commodities markets and the tide of easy money going out has led some, including
renowned economist Satyajit Das, to predict there will now be a shake-out of
"sub-prime oil" companies with many comparing the high-cost, highly geared,
fast-declining shale oil plays in the US to a Ponzi scheme. Debt markets are
already pricing a strong chance the most fragile shale oil players will default
on their borrowings, through trade in credit default swaps. As Satyajit Das told
the ABC's 7.30
at the end of last year, there is a lot of uncertainty about the
sustainability of the US shale oil boom: “We don't know how big these reserves
are, how quickly they deplete and we know that there is a bubble in that because
a lot of money has been borrowed to invest in these.”

Harking back to the sub-prime crisis is highly emotive, but it seems clear
that the sudden oil price collapse is driven by more complex factors than a new
bounty of unconventional oil supply. In its World Energy Outlook, released in
November, the IEA
oil demand would rise by 14 million barrels a day by 2040, from
roughly 90 million barrels a day now. With North American unconventional
production petering out from the 2020s, the IEA said meeting the extra demand by
lifting production above 100 million barrels a day will be a huge challenge,
falling on the shoulders of a very few countries. The peak oil debate is not
actually going away.

In a September piece entitled “Why
Peak-Oil Predictions Haven’t Come True
”, The Wall Street Journal
leaned towards the same conclusion: even if there remains plenty of oil in the
ground, a lot of it will stay there because the cost of getting to it — even
where it is technologically feasible — is just too high. “There has to be a
finite limit” one oil and gas consultant told the Journal: “We face
technical and economic limits more than anything else.”

Those technical and economic limits may be more stretchy than geological
limits, but they are very real nonetheless. Lots of oil will be left in the
ground. The Saudis are the lowest-cost producer — al-Naimi told MEES production
costs were as low as US$4-5 a barrel — but there have long been fears their
official reserve figures are way overstated.

"How much we can trust in what the Saudis are saying is another matter
altogether," said Mushalik. "We are all basically in the dark."

"My personal opinion is that Saudi Arabia — and the statistics show that —
has not increased production and has not contributed to the glut or the
perceived glut in the global oil market. [In] 2014 Saudi Arabia produced just
200,000 barrels a day more than in 2005. In fact I think they are struggling on
that level. They won't exceed it."