Last week we wrote about John Kemp’s column pointing out that CFTC data suggests hedgers — those who are exposed to physical prices through their business operations — fuel resellers and others hedge against their operational exposure to oil prices — collectively had the smallest net short position in six years in late April.
We advanced a few possible explanations of our own, including the idea they’d mainly popped over to the Brent market.
However, was this particular fuss all over nothing? Philip Verleger, a veteran oil watcher and economist, says Kemp is a touch off base.
His first mistake was to rely on the CFTC data at all, says Verleger.
Kemp’s second mistake is to use the net figure for hedgers (aka, “Producer/Merchant/Processor/User”).
To explain, let’s start with Kemp’s opening line:
Oil hedgers held the smallest net short position in WTI-linked futures and options for at least six years in the week ending April 24, according to commitments of traders data published by the U.S. Commodity Futures Trading Commission (CFTC).
To which Verleger says:
That sentence sounds perfectly reasonable. It implies that traders are long 100 and short 100. But wait. Kemp was not talking about speculators, who can hold long or short positions, but hedgers. How can hedgers have a net zero position in futures? The answer is they cannot.
Without belaboring the point, Edwards, Ma, and Williams make clear that the activity of hedgers – the people Kemp described – are either long the physical or short the physical. They are two different groups of traders. Kemp merely observes that the two groups, by random chance, seem to have equal positions.
Verleger has a point here: are the net figures for merchants very meaningful, given that it is the net of the positions of a somewhat diverse group of traders?
Let’s look at Kemp’s graphical representation of the movement of the net positions for each category over the past few years — which does seem to show a trend:
The net position of an entire category seems to intuitively tell us something, when hedgers (green) do seem to be reducing their net shorts shorts — at least in aggregate — and swaps dealers (blue) are increasing theirs.
But it’s complicated by the CFTC’s categories themselves.
Here’s the CFTC’s full definition of “Producer/Merchant/Processor/User”:
A “producer/merchant/processor/user” is an entity that predominantly engages in the production, processing, packing or handling of a physical commodity and uses the futures markets to manage or hedge risks associated with those activities.
Verleger says this means that some in this category will be short (probably producers), others will be long (probably refiners, or Delta ahem) — and not all of this will be captured in the CFTC data. And he adds that the activities of the big integrated producer/refiners — ie, those that are both producers and refiners — probably won’t be represented at all:
If one stops and thinks, the activities of integrated producer-refiners such as ExxonMobil are probably not captured here. Exxon might want to be short crude oil to protect against a decline. However, Exxon’s refining group would want to be long crude oil to protect against a crude price rise while being short products. Any intelligent integrated company – and ExxonMobil clearly passes the test – would understand that it had no reason to sell crude futures short and buy crude futures.
Hence, says Verleger, it’s just “random chance” that the short positions of producers have matched the long positions of crude buyers.
But it really raises the question: what do we know about the parties that hedge physical positions in oil anyway?
As it turns out, the CFTC data combined with data issued by the Intercontinental Exchange tells us a lot. In particular, the net long position of the CFTC merchant category is balanced by a net short position in products. This suggests that refiners are using oil markets to lock in margins, precisely as agricultural processors have used the soybean market for over 100 years. What a surprise!
The two do appear to follow reasonably closely:
One can add that in recent months, gasoline and heating oil stocks have decreased. In fact, heating oil inventories are quite low in Europe and the US. Not surprisingly, merchant short positions in these products have fallen. It should also come as no surprise that long merchants have cut back crude positions.
Verleger’s final point is that Kemp is just looking at WTI data. When Brent is added to the mix, there is not much narrowing of net short positions to be seen.
With just WTI:
With WTI and Brent:
Coincidentally, this was the number one explanation advanced in Izzy’s post last week.
And if you haven’t had enough of this yet, there is one more point that we’d like to return to and emphasise: the CFTC categories themselves remain problematic, even in their “disaggregated” form, which was first published in 2009. The CFTC’s own guide to this data is full of disclaimers, such as (our emphasis):
Some traders being classified in the “swap dealers” category engage in some commercial activities in the physical commodity or have counterparties that do so. Likewise, some traders classified in the “producer/merchant/processor/user” category engage in some swaps activity. Moreover, it has always been true that the staff classifies traders not their trading activity. Staff will generally know, for example, that a trader is a “producer/merchant/processor/user” but we cannot know with certainty that all of that trader’s activity is hedging. Staff is working on improvements to the Form 40 and other methodologies in order to improve the accuracy of the trader classifications.
Kemp in his column mentioned that the CFTC not only doesn’t publish detailed information on how it categorises the traders; it won’t say whether the policy has changed since the disaggregated categories were introduced. Alas, there isn’t much of an alternative.
Where have all the oil hedgers gone? FT Alphaville
There’s always a silver lining, even in a commodities rout – FT Alphaville