Speculating About Oil

ThinkProgress is at it again. Today that web site, which is run by the co-chairman of Barack Obama’s transition team and is essentially a front for the Obama administration, returned to the subject of oil prices. This is an important topic for the Democrats, for obvious reasons: both President Obama and his Secretary of Energy are on record as favoring much higher gasoline prices, which is exactly what their administration has delivered. But Obama has no hope of being re-elected if voters blame him for $4 a gallon gas, so his minions are desperately pointing the finger elsewhere to try to shift responsibility.
Hence today’s story, titled REPORT: How The Kochs Built An Oil Speculation Empire, and the accompanying report, titled “How Koch Became An Oil Speculation Powerhouse.”
The report is the usual left-wing silliness. I asked Craig Pirrong to comment on it. Craig is the author of a book on market manipulation titled The Law, Economics, and Public Policy of Market Power Manipulation, as well as ten scholarly articles on the subject in economics and law journals. His new book, Structural Models of Commodity Price Dynamics, is about to be published by Cambridge University Press. This is part of his response:

I read through that screed. A farrago of disconnected factoids that do not come close to supporting the contentions of the article that (a) Koch is primarily a speculator, and (b) speculation is distorting prices.
One major error–one can use swaps and other financially settled products (i.e., derivative contracts that do not permit the making or taking of delivery, but which have payoffs contingent on the price of energy) to hedge. The article asserts that such instruments are inherently speculative, and the lobbying of the Kochs and others to permit off-exchange trading of such instruments was a nefarious plot to facilitate speculation. This is just plain wrong. …
The subject of the impact of speculation on prices is one of extreme controversy. IMO, there is no reliable evidence that speculation has distorted prices, certainly not by 100 percent. I wrote about the arguments and evidence on this issue in a piece in Regulation Magazine last summer, and analyze the subject in my forthcoming book. The title of the Regulation piece might provide a flavor of its conclusion: “No Theory? No Evidence? No Problem!”
I can say that TP (a very apt acronym) provides zero reliable evidence of a speculative impact. Clip jobs don’t count as evidence. And certainly the TP piece does not present a balanced view of the debate on the subject, which as I noted is quite heated. Reading it would give you the impression that it is an established fact that speculation has distorted prices. This is quite clearly not the case. Not by a long shot.

There is much to ridicule in the ThinkProgress report, and I will return to it for more commentary in days to come. For the moment, I want to focus on one blatant falsehood, which is the foundation of the entire report. Today’s post says:

Even Goldman Sachs concedes that at least $27 of the price of crude this year has been a result of rampant speculation, not supply and demand.

The TP report itself contains an almost identical statement:

Excessive energy speculation today is at its highest levels ever, and even Goldman Sachs now admits that at least $27 of the price of crude oil is a result from reckless speculation rather than market fundamentals of supply and demand.

In support of this claim, both today’s post and the accompanying TP report link to another TP post, dated April 13, 2011, which says:

Saying that “net speculative positions are four times as high as in June 2008,” investment banker Goldman Sachs “issued a warning that the price of oil has grown out of control due to excessive speculation.” The world’s largest commodity trader, Goldman Sachs told its clients that it believed speculators like itself had artificially driven the price of oil at least $20 higher than supply and demand dictate. They even admitted that their work to drive up prices has harmed the American economic recovery, pointing to “nascent signs of oil demand destruction in the US.”

That TP post, in turn, linked to a Wall Street Journal article that quoted from a Goldman report dated April 12, 2011, as well as a Telegraph column that referred to and quoted from the same Goldman document. TP did not link to the Goldman report itself, which was issued by Goldman’s chief oil analyst, David Greely.
Neither the Journal article nor the Telegraph article contains the quote about “excessive speculation.” As best I can tell, ThinkProgress made it up. Neither article says anything similar to TP’s characterization, “Goldman Sachs told its clients that it believed speculators like itself had artificially driven the price of oil at least $20 higher than supply and demand dictate.” Where does that statement come from?
It isn’t credited, but it comes from a Reuters article that referred both to the April 2011 Goldman oil report and to a prior report, also by David Greely, dated March 21. The Reuters article says:

Goldman Sachs (GS.N) rocked oil markets for a second day Tuesday by calling for a nearly $20 fall in Brent crude oil, saying speculators had pushed prices ahead of fundamentals. …
Goldman estimated in a research note on March 21 that every million barrels of oil held by speculators contributed to an 8 to 10 cent per barrel rise in the oil price. …
Using Goldman’s 8- to 10-cent estimates and data on speculators’ positions from the U.S. Commodity Futures Trading Commission, Reuters calculated that as of last Tuesday, the total speculative premium in U.S. crude oil was between $21.40 and $26.75 a barrel, or about a fifth of last Tuesday’s price.

This is the only possible source of TP’s $27 (or “at least $20”) claim. It’s curious, though–why didn’t TP want to link to the Reuters article in support of its assertion? Because Reuters didn’t understand the Goldman calculation, Goldman contacted Reuters and pointed out the error, and Reuters added this correction to its article:

Goldman Sachs disputed the Reuters calculation on speculative premium. The bank clarified that the 8- to 10-cent estimate it provided is only meant to reflect the impact of incremental barrels added — or sold — by speculators in relation to defined events over a given time period. It was not meant to be applied to the total number of speculative positions held as these tend to vary across various CFTC measures of net speculative positions.

So the only support for ThinkProgress’s claim that “speculation” is adding $27 (or “at least $20,” TP offers no explanation for the discrepancy) per barrel to the cost of oil was a misinterpretation of a Goldman Sachs report which was later corrected by Reuters. Would it be possible for anyone who comments on public affairs to be more dishonest than ThinkProgress?
Now let’s look at the two Goldman Sachs reports that are referenced in the linked news stories and, indirectly, by TP. The first is dated March 21, 2011, and as noted above was issued by David Greely. Here is the relevant portion:

While the ongoing economic recovery has now drawn OECD total petroleum inventories back to five-year average levels (Exhibit 1), the oil market remains justifiably focused on the risks to the forward supply-demand balance posed by the dramatic events unfolding across the globe. In addition to the risks to oil supply from the ongoing political unrest in the Middle East and North Africa (MENA) and the impact on oil demand from the damaged nuclear reactors in Japan, this weekend has brought reports suggesting another oil well inthe deep water off the US Gulf Coast is leaking. However, at the time of writing, US Coast Guard states that the observed “oil sheen” is likely the result of sediment and not oil.
In such an environment, it is not surprising that net speculative long positions in WTI crude oil reached a new record high of 391 million barrels. In comparison, when WTI crude oil prices peaked at over $145/bbl in July 2008, the net speculative long position in the light sweet crude oil contract (future and options) was less than 100 million barrels. We estimate that each million barrels of net speculative length tends to add 8-10 cents to the price of a barrel of crude oil. Given that net speculative length has been about 100 million barrels higher since the political protests spread from Tunisia and Egypt to Libya (Exhibit 2), this suggests that the oil market has been pricing a $10/bbl risk premium into the price of crude oil due to concerns over potential political contagion to other oil producing states in the MENA region. This is consistent with the fact that Brent crude oil has been trading near $115/bbl in the recent period, $10/bbl above our 3-month target.

These are the paragraphs that were misinterpreted by Reuters, whose error–after Reuters had corrected it–was repeated by ThinkProgress. But note, also, that what Greely is talking about here is not “speculation,” as that term is normally understood, but rather a “risk premium.” Greely’s point was not that some nefarious speculators were bidding up the price of oil in hopes of cornering the market–or something, leftists like those at ThinkProgress never explain how “speculation” can ever be profitable–but rather that the market was reacting rationally to the threat of supply disruptions. It was the supply side, not the demand side, that was generating the entirely appropriate risk premium.
In short, the March 21 Goldman report refutes, rather than supporting, TP’s claims.
The second relevant Goldman report was dated April 12, 2011. This report made a splash because Goldman, after a long period of being bullish on oil and other commodities, now said that the price risk was neutral. This report is the source of TP’s quote: “Both inventories and spare capacity are much higher now and net speculative positions are four times as high as in June 2008.” But it does not contain the other quotes that are attributed to it by TP. Nothing on “excessive speculation,” nothing to the effect that “speculators like itself had artificially driven the price of oil at least $20 higher than supply and demand dictate.” This is pure fabrication by ThinkProgress.
In fact, an intelligent reader will immediately see that the Goldman report refutes rather than supporting the TP thesis. Again, it is market realities, not speculators’ schemes, that are driving the price of oil:

Net speculative long positions are now four times higher than in June 2008. Concerns about the further spread of protests and revolts to other oil producing countries in the Middle East and North Africa have led to a roughly 100 million barrel increase in net speculative long positions in the oil market since the protests began in late January (see Exhibit 1).
Consequently, we continue to believe that – even with the loss of Libyan production – the oil market has adequate inventory and OPEC spare production capacity to avoid the degree of physical tightness experienced in 2008 well into next year. Although the contagion risk in the Middle East and North Africa (MENA) remains elevated, and the oil market’s ability to weather the loss of supplies from another producer in the region is limited, we believe that – with the market continuing to embed at least a $10/bbl risk premium in prices – the price risk is becoming more symmetric as we believe that the market will experience a substantial correction toward our $105/bbl near-term target for Brent crude oil in coming months.

Again, what we are talking about here is a “risk premium” arising from events in the Middle East–events which presumably were not engineered by American oil companies.
ThinkProgress’s report on oil speculation is riddled with errors, many of them howlers. That web site’s basic problem is that its writers are neither intelligent nor experienced in business. Therefore, they are utterly unqualified to write about topics like commodity pricing. In this instance, however, ThinkProgress’s conduct is worse than incompetent: it released a report that tried to blame “speculators” for the current increase in oil prices. Its only support for that claim consisted of quotes that were falsely attributed to Goldman Sachs. The entire report, therefore, was premised on a lie.

Notice: All comments are subject to moderation. Our comments are intended to be a forum for civil discourse bearing on the subject under discussion. Commenters who stray beyond the bounds of civility or employ what we deem gratuitous vulgarity in a comment — including, but not limited to, “s***,” “f***,” “a*******,” or one of their many variants — will be banned without further notice in the sole discretion of the site moderator.