So the economy is in the doldrums globally and demand for oil languishes. At the same time, oil producing countries are pumping the stuff out as hard as they can with no let-up in sight. So how does it make sense that oil price surged nearly 60% from its January lows? Today’s news and rumors may explain today’s advance, but what about previous three months’ rally? The answer is that over time, it is the price that leads the narrative, not the other way around. I’ll try to elaborate.

During the two decades of my career in commodities trading I have observed time and again how significant price changes shaped the prevailing market narrative. By narrative here I mean a shared interpretation of how key causal forces affect market events. Market fundamentals – the available bits of relevant data – constitute the building blocks that mould our understanding of what’s going on, but what we do with them depends on how we evaluate their relevance and credibility. This is never a straightforward process, so at any one time we can entertain more than one possible interpretation of market conditions.

As a young oil market analyst in the 1990s, I pretty much expected that available fundamentals data gave us a factual account of the world: that “bullish” information would lead to a rise in oil prices and “bearish” information would lead to their decline. Thus, an increase in demand for oil should cause oil prices to rise. So would shortfalls or interruptions to its supply. Conversely, falling demand or increasing production should cause prices to fall. Often however, I observed that price action seemed largely in discord with the fundamentals.

For example, in the late 1990s, global economic growth was in full swing and the demand for oil was rising. Meanwhile, funding for oil production and refining tightened globally as capital favored investments in information and telecommunications technologies. As a result, demand was expected to progressively outstrip supply, pushing oil prices significantly higher in the future. Contrary to those expectations, oil prices more than halved from around $24/barrel in the early 1997 to below $10/barrel in 1999.

Market participants’ need to reconcile the supposedly bullish fundamentals with collapsing oil prices gave rise to stories and rumours about massive stocks of unsold oil and vast tank farms around the world, full to the brim. As prices fell toward $10/barrel, the bearish narrative became entrenched and many traders thought that oil could halve again to $5/barrel. But stories about huge unsold oil inventories (in effect rumours given credence by the declining price) proved unfounded and after bottoming out in 1999 oil prices tripled to $35/barrel over the following 20 months even as the world economy slipped into recession and demand for oil contracted.

Again, the market sought to reconcile these contradictions with a new narrative to fit the events. Now we heard about falling production of oil fields around the world, rising production costs, a shortage of refining capacity and growing demand from emerging economies. One of the biggest stories affecting the market was the peak oil hypothesis. Not that this hypothesis was just then formulated catching everyone by surprise: it was originally advanced by Marion King Hubbert in 1956 and subsequently popularized in the 1970s. Its re-emergence in 2005 and 2006 reflected the markets’ need to explain the oil prices, which continued to break new all-time record highs.

Peak oil and Saudi oil wealth

Peak oil refers to the point in time when worldwide oil production passes its maximum point, followed by an irreversible decline. According to various interpretations, this may have already happened between 2005 and 2009[1]. Given the massive relevance of this hypothesis to an oil trading firm, I made a concerted effort to get to the bottom of the issue. I expected to unearth the truth of the matter. Instead, I encountered widely diverging views and dissonant information produced by different agencies and research outfits.

In particular, there was a stark contrast between the views espoused by proponents of the peak oil hypothesis and the conventional view of the market held by the industry[2]. Peak oil researchers held that we are entering a period of terminal decline in oil production and that oil prices will get much, much higher in the future. The industry view held that crude oil was very plentiful around the world and that new deposit discoveries and improved drilling technologies would keep the world abundantly supplied at stable prices for decades.

Happy talk about plentiful oil usually invoked Saudi Arabia’s vast reserves and production capacity. For years, the kingdom was believed to have some 260 billion barrels of proven oil reserves together with another 200 billion of probable reserves. It had not occurred to me to question these figures until I started to scratch a bit below the surface. The magic of Saudi oil reserves was that they kept constant (or even increased) in spite of the extraction of close to 3 billion barrels each year.

After twenty years of that, you’d think that reserves would decline by 50 or 60 billion barrels. But no: by 2014, Saudi Aramco claimed that they had 790 billion barrels of oil resources and expected this figure to hit 900 billion barrels by 2025[3]. This bonanza did not come about from discoveries of giant new deposits[4] but from the changing definitions of oil and from a subtle shift in terminology.

While most of the press uses the terms reserves and resources interchangeably, it is very important to discern between the two. Resources comprise oil from contingent and prospective sources which include quantities that are potentially recoverable from accumulations that are as of yet undiscovered[5]. Thus, oil resources are by definition wide open to exaggeration and wishful thinking. What we have traditionally understood as “reserves,” usually represents only a small fraction of resources that can be feasibly developed.

If we revert to the traditional Proven Reserves Method[6], Saudi reserves appear much less abundant. The last audit of Saudi reserves complying with this methodology was done in 1979 and showed that Saudi Arabia had 110 billion barrels of proven reserves, another 67 billion barrels of probable reserves and 69 billion barrels of possible reserves (reserves are classified as proven if there is 90% confidence of them being recoverable with existing technology and under current economic and political conditions; they are probable if there’s a 50% confidence of them being recoverable; for possible reserves, there has to be at least 10% confidence of recoverability under existing circumstances).

Given that 100 billion barrels have already been extracted between 1979 and 2015[7], Saudi Arabia appeared dangerously close to running dry. Work of peak oil researchers like Matthew Simmons, Collin Campbell and Michael Ruppert corroborated this scenario as did the leak of 2007 confidential U.S. Embassy cables from Riyadh published by the Guardian newspaper[8]. Such information was never mentioned in the industry publications and only very exceptionally by the mainstream press.

It seemed to me that, between the conflicting figures and narratives, arriving at an objectively correct projection of future trends in the oil industry was quite out of the question. This research dispelled my illusion that diligent research of supply and demand fundamentals could conceivably lead to more reliable forecasting of the future price of oil or any other asset for that matter.

There was no reason to believe that the information on other industries was any better. Take the example of South Africa’s gold reserves. For decades, South Africa had been one of the world’s largest producers of gold. According to a revision in 2001, their gold reserves were pegged at 36,000 tons of the precious metal, about 40% of the world’s total. However, United States Geological Survey subsequently estimated that South Africa only had 6,000 tons worth of feasibly extractable gold reserves left. Later research by Chris Hartnady of the University of Cape Town showed that the country’s true reserves were perhaps as low as 3,000 tons.

The discrepancy between 36,000 and 3,000 tons again puts the whole way we obtain such information in doubt. Neatly tabulated figures published in serious looking research reports had a feel of factual truth, yet I couldn’t help but wonder how those figures came about. In his book, “The Lexus and the Olive Tree”, Thomas Friedman explains how he filed temperature reports for Beirut when working there as a correspondent for the New York Times. “I estimated what the temperature was, often by ad hoc polling,” writes Friedman. “Gathering the weather report basically involved my shouting down the hall or across the room: ‘Hey, Ahmed, how does it feel out there today?’ And Ahmed or Sonia or Daoud would shout back, ‘Ya’ani, it feels hot.’ … So I would write, ‘High 90 degrees.’”^[9] Friedman’s reports were then duly included in UPI worldwide report from Beirut.

Once published in reputable newspapers as the New York Times or the Washington Post, they appeared as facts, black-on-white, but as Friedman confesses, they were merely his own lazy guesstimates. I wondered how much of the information presented in the compelling research reports I occupied myself studying, came from surveys conducted with similar rigour. Once they were cited by respected institutions however, they gained the validity of hard facts, giving us the sense that we could understand what’s going on in the world and why. I was beginning to suspect that often we didn’t.

Alex Krainer is an author and hedge fund manager based in Monaco. Recently he has published the book “Mastering Uncertainty in Commodities Trading“.

Notes:

[1] In its 2010 International Energy Outlook, the U.S. Energy Information Administration (EIA) proclaimed that oil production from conventional sources probably peaked in 2006.

[2] By industry, I mean the oil corporations, their bankers and a myriad consultancies and analysts.

[3] Reuters: “Saudi Aramco’s Oil Resources to Grow to 900 bn Barrels by 2025.” 19 Nov.2014 – http://gulfbusiness.com/2014/11/saudi-aramcos-oil-resources-grow-900bn-barrels-2025/#.VONRFSztCYM

[4] The last great Saudi oil field was discovered in 1967. To date, only smaller deposits have been found and the bulk of them have not yet been developed.

[5] “Recoverable” doesn’t necessarily mean “economically recoverable,” which would imply that the value of extracted oil should cover the costs of exploration, drilling, extraction, transportation and a certain return on invested capital.

[6] This methodology was required by the U.S. Securities and Exchange Commission, but was last performed on Saudi Aramco’s reserves in 1979. After the control of Saudi Aramco passed from American management to the Saudi Petroleum Ministry no further surveys using this methodology have been conducted.

[7] Saudi oil production by year: http://www.indexmundi.com/energy.aspx?country=sa&product=oil&graph=production – according to the U.S. Energy Information Administration figures, Saudi Arabia extracted 96.21 billion barrels from 1980 through 2013. At 9.5 million barrels per day, the total through 2014 rises to 99.68 billion barrels of already extracted oil.

[8] In 2010 Wikileaks released confidential U.S. Embassy cables from Riyadh that were then published by The Guardian newspaper. One of the cables from 2007 recapitulated U.S. Consul General’s meeting with Mr. Sadad al-Husseini, Aramco’s Executive Vice President for Exploration from 1992 to 2004. According to this cable, Mr. Husseini asserted that at that time, Saudi Arabia had 64 billion barrels of remaining oil reserves and that these reserves would last 14 years (i.e. until 2021), after which Saudi output would enter a period of steady decline that no amount of effort would be able to stop. A different report by Citigroup in 2012 further confirmed the dire situation with Saudi oil reserves concluding that failing to discover major new oil fields, the kingdom was liable to cease exporting oil altogether by 2030.

[9] Friedman, Thomas. “The Lexus and the Olive Tree.” New York, Anchor Books, 2000.