COLUMN-Consolidated analysis of oil positions needed: John Kemp
— John Kemp is a Reuters market analyst. The views expressed are his own —
By John Kemp
LONDON, May 25 (Reuters) – The standard analysis of speculators’ positions in crude oil markets is highly misleading. By focusing only on futures and options positions in physically-settled NYMEX light sweet oil it ignores important and financially equivalent positions in other WTI-linked derivatives as well as positions on the rival ICE market in London.
The anomaly has been thrown into stark relief by proposals published by the Commodity Futures Trading Commission (CFTC) earlier this year that would start applying position limits for referenced energy contracts on an aggregated basis across similar commodities.
In addition to the four referenced contracts (NYMEX crude, gasoline, heating oil and natural gas) aggregate limits would apply to any other contract exclusively or partially based on the same commodities and deliverable at the same location (75 Fed Reg pages 4144-4172).
While other aspects of the plan have attracted fierce criticism, most accept the logic of applying some form of aggregation when trading is split across different but financially equivalent contracts.
The CFTC plan has highlighted shortcomings in the way in which industry analysts themselves calculate positions. For the market, better aggregation across financially equivalent contracts is a much-needed first step to improved understanding of how investors, swap dealers and commercial hedgers use the market and what impact they have on both prices and the shape of the forward curve.
Analysis of hedgers’ and investors’ positions is usually presented something like Chart 1 (see the attached chartbook using the link below).
Chart 1 shows the net long (above the line) or net short (below the line) position of commercial and non-commercial traders in physically settled NYMEX light sweet oil. It is based on weekly data from the CFTC’s commitments of traders (COT) report. Options are shown on a futures-equivalent basis by adjusting them using market deltas.
While it is widely used, the traditional analysis has limited value. By definition commercial and non-commercial positions must be equal and opposite. The balance is held by non-reporting entities, is usually small, and often simply lumped in with the other non-commercials.
Any increase (decrease) in the non-commercial net long must be offset by an equivalent decrease (increase) in commercial hedgers’ net short. So it is not clear what significance (if any) should attach to a change in the non-commercial net long. Who moves first – do changes in commercial hedgers’ positions force non-commercials to respond? Or is it the other way around?
The CFTC’s own economists concluded commercial hedgers normally initiate position changes. Hedge funds (“money market traders”) and other non-commercials mostly provide liquidity by reacting to hedgers’ moves, taking the opposite side of the market (“Price Dynamics, Price Discovery and Large Futures Trader Interactions in the Energy Complex”, CFTC, 2005).
But the study looked at behaviour prior to 2005. It is not clear whether this pattern of hedgers being active and non-hedgers being passive still holds. Non-hedgers now account for a much larger share of the market than before 2005.
Historical commitments of traders data has been criticised on other grounds. Positions are classified by the predominant nature of the owners’ business (commercial or non-commercial) rather than the nature of each position itself. For example, if an oil company holds some positions as inventory hedges and others as part of a speculative trading book, all positions are reported as “commercial” in the weekly COT reports.
The other criticism is that binary distinction between commercial hedgers and non-commercial speculators or investors is far too simplistic and offers too little detail, especially about the make-up of each group.
The “commercial” category includes many swap dealers who use the public markets to “hedge” OTC swaps and “financial” exposures with pension funds and commodity merchants, rather than hedge the physical commodity itself. Critics charge that swap dealers have more in common with non-commercial investors than with commercial hedgers, and are misclassified.
The CFTC has addressed at least some of these criticisms by making more detailed data available. It now classifies positions in five separate categories (producers, consumers, merchants and manufacturers; swap dealers; managed money; other reporting; and non-reporting). While this is still not as detailed as the private breakdown which the CFTC gets as part of its Large Trader Reporting System, it is a marked improvement on the previous system.
The disaggregated numbers paint a more nuanced picture (Chart 2). Producers, consumers and merchants have run a net short position continuously for the last five years, offset by a net long position by the managed money community (hedge funds, exchange traded funds, and commodity pool operators). Swap dealers and other (small) non-reporting traders “balance the market”, running mostly net long positions but occasionally appearing on the short side.
Even the disaggregated analysis is incomplete and potentially misleading. It focuses on just one set of derivative contracts (NYMEX physically delivered light sweet crude futures and options). In reality there are now many others which are also priced on substantially the same commodity at the same location and which are therefore financially and economically equivalent.
For light sweet crude deliverable at Cushing (Oklahoma) other contracts include financially settled light sweet oil (WS), average price options (AO), calendar swap futures (CS) and WTI-lookalike or European style options (LC). The rival Intercontinental Exchange (ICE) offers its own financially settled WTI lookalike contract (WTCL).
In fact the main NYMEX WTI futures (CL) and options (LO) now account for just 60 percent of all open interest in WTI-linked derivatives. The balance is in other option and swaps contracts offered by NYMEX (30 percent) or in WTI-based products offered by ICE (10 percent) (Table 1).
“Historically, all trading volume in a specific contract tended to migrate to a single contract on a single exchange”, according to the CFTC. “With the advent of lookalike energy contracts that are listed on different registered entities and contracts that are based on other contracts in an attempt to isolate different energy price risks (failure to aggregate across contracts) could result in applying a position limit to only a very limited segment of a broader regulated market”.
Crucially, positions in these other contracts do not follow the same pattern as in the main physically settled WTI futures and options. In particular, swap dealers tend to be short of both averaging pricing options and European style lookalike options, hedging their price exposure in the (more liquid) main futures and options market (see Charts 3-7).
Any serious analysis needs to take into account positions in these other derivatives as well as in the main NYMEX WTI futures and options.
Chart 8 provides a first attempt to bring all the main WTI-linked positions together in one aggregate. Banks and other swap dealers play a much more important market balancing role. They are more often net short in other contracts than in the main WTI futures and options, and there position varies more frequently from net long to net short overall.
Note how swap dealers’ net position across all WTI-linked contracts turned short in Q2 2008 (before prices peaked at $147 in July 2008) and again in Q1 2010 (as the market was surging back towards $87.
As trading fragments across an increasing number of contracts and exchanges, useful analysis needs to move beyond the conventional NYMEX net long/short graphs to take an aggregated look at positioning across all related markets.
((firstname.lastname@example.org; reuters messaging: email@example.com; +44 207 542 9726))