Monday, August 31, 2015

What Do Futures Markets Tell Us About Long-term Oil Prices?

  • The tendency to believe that the prices of oil futures contracts are predicting the future price of oil is understandable but not supported by the track record of such bets.
  • The prices of long-dated oil futures merely reflect where buyers and sellers are willing to strike a deal today, for their own, diverse reasons.
A recent article in the Wall Street Journal reminded me of numerous debates about the significance of energy futures prices, when I was a trader and later a trading manager for the former Texaco, Inc.  Do changes in futures contract prices actually predict future oil prices as the Journal's reporter suggests? If so, then it might be reasonable to conclude that today's low oil prices could persist for years. However, from my perspective that over-interprets the market data and ignores some important oil fundamentals.

As tempting as it might be to think so, the futures market for West Texas Intermediate (WTI) crude oil isn't a crystal ball, and neither is the market for UK Brent crude. A futures price is simply the price someone is willing to pay or receive now for oil to be delivered (or settled without delivery) later. It is typically based on business needs, rather than deep analysis.  A concrete example might be helpful.

The parties who on August 11th bought or sold oil for $56 or $57 in December 2017 likely did so, not because they were certain what the price would be then, but because they couldn't be sure and either needed to hedge another transaction or activity, or thought it constituted a reasonable bet. Aggregating a modest number of such transactions--long-dated futures trade much less frequently than those for the near months--doesn't improve the accuracy of these bets on an inherently unpredictable commodity over long intervals. Anyone who thinks it does should examine the track record of oil futures as predictions; it is a sobering exercise, especially for those who have traded this market.

Consider that while the September 2015 WTI contract closed at a little over $43 per barrel that afternoon, traders were buying and selling the same contract for more than twice as much during long stretches of 2012--about as far removed from us as the late-2017 contract prices cited in the Journal article as evidence of a persistent oil-price slump. Prices for the September 2015 contract were even higher in the middle of last year, when traders knew nearly as much about the growth of US tight oil production and its rising productivity as we do today, but crucially didn't know that OPEC would choose not to cut output to alleviate an over-supplied market as they had done in the early 1980s and late 1990s. Similar examples abound.

So how else might one explain the fact that long-dated oil contracts are trading for less today than they were this spring, if not as a prediction of a longer period of low prices ahead? Behavior and learning play key roles. With the  first anniversary of this historic price collapse just a few months off, expectations of a quick rebound in prices have faded. The possibility that the US could produce as much tight oil, for now, with fewer than half as many drilling rigs in operation as a year ago has sunk in. So has the reality that as painful as $50 oil is for some of OPEC's members, cartel leaders like Saudi Arabia show little inclination to blink first.

However, others are blinking, and that's why I'm skeptical that oil prices can remain this low indefinitely. The cuts in staff and investment budgets by major oil companies and their national oil company peers have been breathtaking, totaling $180 billion this year according to one analysis. The cuts suggest that the projects in question require significantly higher oil prices to be profitable, even after recent cost reductions, or have become too risky at current prices.

Few of these companies are big players in shale. Their bread and butter is large, conventional onshore oil fields and enormously expensive deepwater oil projects, the collective output of which is inherently subject to annual declines in output. Decline is the "silent killer" of output, to the tune of 5% or so every year. The only way to offset this trend within the portfolios of these producers is to spend large sums every year on new wells and new projects--projects that according to Rystad Energy, as cited by Bloomberg, have been cut more than at any time since 1986.

We must also put the US shale revolution in its proper context. When added to a global market that was balanced between supply and demand at around $100 per barrel, it was a game-changer, not least because no other producer or group of producers was willing to reduce output enough to accommodate this new source. However, even at today's 5.4 million barrels per day US tight oil represents only about 6% of global supply. The combination of shale plus OPEC covers less than half the world's oil demand.

The remainder must come from onshore and offshore oil fields in non-OPEC countries like Brazil, Canada, Mexico, Norway, and Russia. This non-OPEC supply has grown thanks to  a wave of completions of  large projects begun 5-10 years ago, when prices were rising rapidly. However, reduced investment now surely means lower non-OPEC production within a year or two.

The key question for future oil prices is therefore when demand, which according to the International Energy Agency is growing rapidly under low prices, and supply, for which new investment has suddenly shifted from the accelerator to the brake pedal, will cross over, erasing today's glut. It's hard to infer the answer from the thinly traded market for long-dated oil futures contracts.


Nony said...

The first point is just one of mean estimate versus variability. EIA publishes a product that shows the implied 90% confidence interval of future price, based on options. It is wide.

However, the central estimate is still interesting, especially when it moves significantly. The few year out estimate actually moves less than the next month does. Is less sensitive to near term disruptions, weather, macro cycles, etc.

e3 said...

Now that more people are opting for alternative energy sources, such as solar energy and wind power, it will be interesting to see how this market fares over the next decade or two.