Home Energy and Environment Goldman Sachs Nailed Oil Price “Super Spike”; Other Forecasters Not So Much

Goldman Sachs Nailed Oil Price “Super Spike”; Other Forecasters Not So Much


Over the first 3 1/2 months of 2011, according to EIA, the price of (benchmark West Texas Intermediate) crude oil in the United States has averaged around $96 per barrel. This compares to $61.95 and $79.48 per barrel in 2009 and 2010, respectively, during a severe economic downturn that caused world oil demand to fall sharply from pre-recession levels/growth rates. Today, with the world economy – and, hence, world oil demand – rebounding, combined with unrest in the Middle East and an oil supply disruption in Libya, oil prices (and, of course, gasoline prices) have moved back up as well.

Now, none of this should be particularly surprising to oil market economists, but I was curious — did any of them actually manage to correctly forecast this situation?  First, let’s start with one outfit that actually suspected something like this might happen, back in March 2005 no less!

Oil prices have entered the early stages of a multi-year period of trading in which economic growth and rising demand could push oil to $105 per barrel, enough to meaningfully reduce energy consumption, Goldman Sachs analysts said Thursday.

We believe oil markets may have entered the early stages of what we have referred to as a “super spike” period — a multi-year trading band of oil prices high enough to meaningfully reduce energy consumption and recreate a spare capacity cushion only after which will lower energy prices return,” said analyst Arjun Murti.

When this forecast came out, I was working as a world oil markets expert at EIA. I clearly remember most people scoffing at the Goldman Sachs forecast (note: not to pat myself on the back, but along with a few others at EIA, I was very bullish on oil prices at that point — based on rapidly growing world oil demand and tightening OPEC spare capacity, among other factors — and most definitely did NOT scoff at Murti). In fact, as this article explains, “When Murti first predicted oil could reach a $105 a barrel, way back in March of 2005, oil was trading below $60. The prediction was scorned and laughed at, even as it caused a bump in crude prices… some thought it was just a Goldman Sachs publicity stunt.” Instead, as it turns out, “Murti’s call was dead right, and all the naysayers were wrong.

Other than Murti, what were most other forecasters expecting oil prices to be? I looked back at past long-term outlooks from EIA and other analysts, and the bottom line is that almost nobody saw this coming. For instance, back in 1998-2000, major forecasters (e.g., the IEA, EIA, DRI, PEL, PIRA, and GRI) were all predicting oil prices for 2011 (I’ve extrapolated between 2010 and 2015, since there are generally no point estimates for 2011 per se) under $30 per barrel (in $2010). Clearly, nobody was even close to where we’re actually at today — nearly $100 per barrel for the first 3 1/2 months of 2011.

So, I figured that as the forecasts got closer in time to today, they’d get significantly better. But no…not particularly. For instance, forecasts for 2011 made in 2004-2005 were generally in the upper $20s per barrel (in $2010), to about $40 per barrel in the case of PIRA. Again, not even close to where we’re actually at.

Finally, in 2007-2008, with oil prices running up sharply, forecasters started adjusting, but generally they lagged consistently – and significantly – behind reality. For instance, the IEA’s 2007 forecast was for oil prices in the mid-$60’s per barrel range for 2011. Closer…but certainly no cigar. Same thing for EIA. As for CERA and its famed founder, Daniel “The Prize” Yergin, they weren’t even close. For instance, in this mid-2007 interview, Yergin noted that CERA’s long-term forecast expected oil prices to fall sharply from then-prevailing levels (around $65-$70 per barrel). Uhhhhh…don’t think so.

The bottom line? Almost nobody, with the notable exception of Goldman Sachs and its then-controversial “Super Spike” forecast, even came close to nailing where oil prices would end up in early 2011. My theories as to why that is the case?

First, it’s almost impossible to forecast economic ups and downs, such as the financial meltdown of 2008. If you can’t get the economics right, then it’s almost impossible to get oil demand right either. And, since skryocketing world oil demand – in China, India, Brazil, etc. – has been the big driver of oil prices in recent years, there go the forecasts…down the toilet.

Second, it’s almost impossible to predict political unrest, wars, and oil supply disruptions. Yet those matter a great deal. Today, for instance, there’s almost certainly a significant “risk premium” built into oil prices due to Middle East unrest (and concerns about said unrest spreading to “The Prize” – Saudi Arabia and the Persian Gulf region in general), although I doubt that explains more than $10-$20 per barrel of current world oil prices. In addition, there’s an actual oil supply disruption in Libya, although we’re only talking about a net supply disruption of perhaps 0.5-1.0 million barrels per day, and only for a fairly short period of time so far. For both of those reasons, I strongly doubt that the Libyan disruption could account for more than $5 per barrel or so of current world oil prices (note: EIA’s “rule of thumb” is that for every 1 million barrels per day disrupted, we can expect oil prices to rise about $5-$10 per barrel or so).

Third, there’s a tendency among forecasters of all kinds to stick with the herd, and also with the price trend. Very few forecasters, except for the crazy and courageous, are willing to stick their necks out on the line, and when they do – as Arjun Murti found out in 2005 – they get roughly slapped back into line.

Finally, there’s also a strong tendency among oil price forecasters to assume abundant supplies of oil well into the future, certainly no “peak oil,” and in cases like CERA, an extremely bullish outlook on oil production for many years to come. When you consider all those issues, it’s a wonder that any oil analyst ever gets their price forecast right (let alone being “right for the right reasons!”). What’s even more a wonder, given their generally pathetic track record, is that people keep believing oil price forecasts. But, then again, humans have always had a burning, insatiable desire to “know” what’s going to happen in the future, and if they need to read cat entrails or use Ouija boards — or complex econometric models that almost nobody can really understand, but that ultimately prove the maxim, “garbage in/garbage out”) — so be it.

P.S. h/t to The Oil Drum for inspiring me to want to write this piece.

P.P.S. Needless to say, today’s oil prices have next to nothing to do with any energy policies enacted in the last year or two. It simply takes far longer than a year or two for any significant additions (or policy-driven reductions) to energy supplies, so that’s just not plausible. On the other hand, policies which encouraged our economy to recover and which prevented a world collapse most definitely resulted in higher oil demand, ceteris parabis, and thus higher prices than would have been the case. Personally, I’ll take the economic recovery and higher oil prices any day over the alternative scenario.

  • jack russell

    From here:


    3.10 Wicked Problems

    Nancy Roberts of the Naval Postgraduate School has studied the types of problems that the U.S. government has faced in recent conflicts:

    Government officials and public managers are encountering a class of problems that defy solution, even with our most sophisticated analytical tools. These problems are called “wicked” because they have the following characteristics:

    1. There is no definitive statement of the problem; in fact, there is broad disagreement on what “the problem” is.

    2. Without a definitive statement of the problem, the search for solutions is open ended. Stakeholders-those who have a stake in the problem and its solution-champion alternative solutions and compete with one another to frame “the problem” in a way that directly connects their preferred solution and their preferred problem definition.

    3. The problem solving process is complex because constraints, such as resources and political ramifications, are constantly changing.

    4. Constraints also change because they are generated by numerous interested parties who “come and go, change their minds, fail to communicate, or otherwise change the rules by which the problem must be solved.”67

    Roberts distinguishes wicked problems from other types of problems:

    Type 1 problems, or what I call “simple problems,” enjoy a consensus on a problem definition and solution. For example, a group of machinists agree that a machine has broken down and they also agree how to fix it. Problem solving is straightforward engendering little if any conflict among those involved. Given their training and experience, these problem solvers, within a short period of time, recognize what the problem is and activate established routines and standard procedures to deal with it.

    Type 2 problems introduce conflict to the problem solving process. I call them “complex problems”. Although problem solvers agree on what the problem is, there is no consensus on how to solve it.

    Type 3 problems engender a high level of conflict among the stakeholders. In this instance, there is no agreement on the problem or its solution. Nothing really bounds the problem solving process-it is experienced as ambiguous, fluid, complex, political, and frustrating as hell. In short, it is wicked.68 Many problems faced by the interagency team, particularly those relating to irregular threats, fall into this third category. Depending on their perspectives, each team member may view an irregular threat in a different light and have a different view of the solution.

    An attempt to leap into the solution space without first agreeing on the problem can lead to the kind of disjointed approach observed by the Special Inspector General for Iraq Reconstruction (SIGIR): “the [Coalition Provisional Authority] approach to reconstruction [had] a disjointed and ad hoc quality: Get the oil flowing. Stop the smugglers. Get the electricity up and running. Clean out the sewers. Rewrite the textbooks. Change the currency. Employ more Iraqis. Focus on the cities. Focus on agriculture. Focus on security.”69

    I would argue that both peak oil and climate change fall into the category of “wicked problems”.