In a recent edition of his widely-followed newsletter (The Gartman Letter, 7 February 2017), commodities king Dennis Gartman observed that, “As of mid-week last week, the hedge fund community was long nearly 1.0 billion barrels of crude while short only a bit more than 100 million barrels. This is the most decidedly one-sided position held by the funds since the late winter of ’14.”
Why is informed money so long and what are the fundamental grounds for all this institutional investor enthusiasm?
In an overview of the supply/demand dynamics of the crude oil market, Nick Cunningham highlighted not only the glut in WTI, but also the product glut. “Oil and refined products inventories in the U.S. continue to climb at a worrying pace, raising some red flags for an oil market that was supposed to be on the mend.” The factors being discussed here include:
• Crude oil inventories in the U.S. now exceed 500 million barrels, having risen every week of 2017, and are not far off their 80-year highs reached in 2016.
• Gasoline inventories are also rising significantly and now stand at over 260 million barrels.
• The rig count in the U.S. is at just under 600 and U.S. production is at just under 9 million bpd – both clearly on the rise.
• Demand in the U.S., Europe, India and China is far from robust based on anecdotal reports from a variety of sources, including shipping brokers and futures merchants.
• The OPEC freeze has come at a time of the normal refinery maintenance season in the Mideast. Even if OPEC has succeeded in reducing supply by one million or so barrels per day, that’s still only a little over 1 percent of daily demand, and it can swing the other way the moment the Saudis or anyone else wishes it to. Related: How Long Can The Permian Craze Continue?
Nick concludes, “The sudden and sharp increase in both crude oil and refined product stocks is a warning sign for oil traders that have by and large been betting on a tightening market and rising prices.”
So What’s the Disconnect?
In a Reuters interview published in March 2011, Craig Donohue, then chief executive of the CME, parent of the NYMEX, commenting on the amount of machine-driven trading in the crude oil futures market said, “As a rough guide 45 percent is proprietary electronic trading and a smaller percentage of that is true high frequency algorithmic trading,” adding that automated volumes on equities were around 30 percent higher.
Today, six years on, estimates vary, but that number is considerably higher. On an asset-class by asset-class basis, machine-driven trading accounts for far and away the bulk of trading in major public markets today, from crude to equities.
In a recent publication from the members-only LinkedIn group, Algorithmic Traders Association, published November 28, 2016, Jonathan Kinlay, PhD, Head of Research & Trading at Systematic Strategies LLC JonathanAlgorithmic Trading, provided a plain-English explanation of machine-driven trading and in particular how computers can be taught to “read” the financial press just as we normal humans do. I’ll abridge his comments: “Text and sentiment analysis has become a very popular topic in quantitative research over the last decade…In the early days, the supply of machine readable content was limited to mainstream providers of financial news such as Reuters or Bloomberg. Over time this has changed with the entry of new competitors in the provision of machine-readable news, including, for example, Ravenpack or more recent arrivals like Accern. Providers often seek to sell not only the raw newsfeed service, but also their own proprietary sentiment indicators that are claimed to provide additional insight into how individual stocks, market sectors or the overall market are likely to react to news.” Related: Trump Burning Bridges In Iraq Over “Take The Oil” Comments
The key point is that financial journalistic verbiage is stripped down to key words or sentiment indicators, words or phrases like “successful,” “market discipline,” “supply reduction,” “exceeding targets,” etc. , and the so-called news-trading algorithm is based on the distribution curve of such sentiment. Orders to the futures pits follow. Doubtless this is a gross simplification, and Jonathan’s article bears reading for anyone not aware of the pervasiveness of machine-driven trading.
But the critical point is this: did anybody say anything about supply and demand here?
If machines really are the new market-makers in the crude oil futures market, then the prices they are spewing out may be as much a mirage as that found when travelling in the OPEC desert of misinformation and fake news.
By Brian Noble for Oilprice.com
More Top Reads From Oilprice.com: