Reader Questions Answered

Futures vs. ETFs Performance Differences

Here’s an interesting, but somewhat complex question about oil trading I had come in by email. Well, at least the answer is complex.

Hi John,

While checking a performance comparison for the last 200 days between USO and the WTI Continuous Contract, I noticed that they had very similar performance up to the end of 2008, and from the start of 2009 they began diverging, with the ETF significantly underperforming the futures contract. What happened around new year’s time that suddenly caused this divergence? And what other ETF can I choose that better mimics the price action of WTI if I would rather avoid trading the futures contract?

Thank you


I’m not an oil trader or analyst (though back in the day I would cover the energy markets on a fill-in basis from time to time), so the really market-specific stuff is beyond my knowledge. I’ll do my best, though, to explain what I suspect is contributing to the difference in performance between the USO and the front month futures, which is basically what the continuous contract tracks.

US Oil Fund (USO) ETF
First of all, we need to look at what makes up the holdings of the USO. According to the fund website, the objective of the fund is:

… for the changes in percentage terms of its units’ net asset value (“NAV”) to reflect the changes in percentage terms of the spot price of light, sweet crude oil delivered to Cushing, Oklahoma, as measured by the changes in the price of the futures contract for light, sweet crude oil traded on the New York Mercantile Exchange (the “NYMEX”), less USO’s expenses.

At this writing the fund is indicated to be long about 45,000 contracts in the July futures, a position with a value of about $2.79 billion as of the May 20 closing price. The fund also shows nearly as much in cash ($2.76 billion). In and of itself, that sort of positioning would be a very good reason why USO is under-performing the front month futures contract. The ETF isn’t fully exposed to it so they cannot possibly match the gains.

I don’t know how those relative weightings have changed over time, though, so I cannot go beyond the current observation to specifically say that’s the entire reason. There may be other factors involved as well (see below). The big cash position definitely stands out, though.

Other Oil ETFs
There are a handful of other oil-focused ETFs out there. Here’s a list:

  • iPath S&P GSCI Oil Total Return ETN (OIL)
  • PowerShares DB Oil Fund (DBO)
  • United States 12 Month Oil Fund (USL)
  • PowerShares DB Crude Oil Long ETN (OLO)
  • United States Heating Oil Fund (UHN)

Some of the ETFs take an approach similar to USO in which they focus their holdings on front month futures to try to track nearby prices. Others, though, such as USL, actually use a spread of futures contracts across an array of months. That creates an interesting dynamic when it comes to comparing performance vs. the front month futures.

Contango and Backwardation
In normal circumstances prices of futures further into the future are priced cheaper than those in the front month. That’s a situation known as backwardation. The price of oil in the ground is less than the price of one in a barrel. Sometimes, though, things flip around into contango, which is when foward prices are higher than current ones.  That can come when near-term demand drops or there’s a glut of supply.

Now, if things hold in either contango or backwardation one would generally not expect to see much difference between the performance of front month oil and that of an ETF which includes holdings of several different month futures. When the situation flips, though, one of the two is going to outperform, potentially by a large margin. For example, if the market goes from backwardation to contango (means forward contracts go from priced less than front month to being price higher) an ETF holding forward months is going to outperform the front month futures (and vice versa in a contango to backwardation switch).

So the bottom line in all this is that you should make sure you know what the ETF you’re trading or investing in is holding so you can know what to expect.

By John

Author of The Essentials of Trading

9 replies on “Futures vs. ETFs Performance Differences”

John, there are two problems to your argument.

1. 45,000 contracts of front month WTI Crude is not worth 2.79 million as you have outlined. With a per barrel price on oil at an estimated / example price of $60 the contracts are worth $2.7 BILLION and not million. 1 WTI Futures contract on the Nymex is equivalent to holding 1,000 barrels of oil (Cited: Therefore the calculation should be 45,000(Total Futures Contracts) x 1,000(# of barrels per contract) x $60(Estimated current price of oil). You will also see that the funds net assets as of 3/31/2009 are 2.93 Billion if they only held 2.79 million in oil, and 2.76 million in cash, the company would have been stealing over 2.8 Billion dollars.

2. Contango is the real reason why there is under performance in ETFs that use front month futures because of the rolling of the contract. I will explain…As an exaggerated example, if July 2009 oil is at $60 and September 2009 Oil is at $100 (Once again exaggerated example) and USO owns 45,000 July 2009 Oil at $60. When July, 2009 comes around and the contract expires the USO fund must then purchase $100 September oil in order to continue to track the index. The rolling of contracts becomes the major issue for all of these commodity ETFs. It can either be in your favor during backwardation or out of your favor during Contango.

Kevin H – A little error there with the m’s and b’s. You’re right and I’ve corrected. The relative ratio is the same, however. This is according to the ETF’s website. As for your second point, duc beat you to it with his link. Thanks for laying it out though. Like I said in the post, I’m no expert on the oil market. Part of why I post these sorts of things is to get the input from those who know more about a subject than I.

John, the reason for the ratio between cash and the value of the position is due to margin. On futures contracts the margin requirement for 1 contract is $5750 ( Therefore, to hold 45,000 contracts you only need to put up 225 million dollars. The rest of the 2.93 Billion sits in cash, tips, or some other type up short term money market fund. Commodities ETFs have 3 parts to their performance.

1. The change in value of the underlying front month future.
2. The change in value of the collateralized portion (Cash, TIPS, Treasuries)
3. The change in value of the effect of backwardation or contango

Put all of those together and you have your performance. Does that make sense?

Kevin – I get all the points your making. It does not change my point, though. How can the fund expect to match the performance of the front futures (the stated aim) when it’s market exposure is so low relative to its asset base. Certainly the managers aren’t looking for the cash and related investments to bridge the gap. If so they might as well make it a fixed income fund.

John, its market exposure is not low relative to its asset base. It’s asset base is 2.93 Billion and its exposure to front month crude is 2.93 billion. The fact that they have 2.73 Billion in cash is a moot point b/c futures are leveraged positions. They control 2.93 Billion of front month crude and only have to put up 225 million dollars for margin purposes therefore they still have 2.73 billion in cash. Here is how they control 2.93 Billion of front month crude.

Buying 1 Futures contract is equivalent to the exposure of 1,000 barrels of oil or approximately $60,000 (using an oil price at $60) but it is only necessary to put $5,750 of your cash out of pocket for initial margin requirements. You therefore get the return of 1,000 barrels or $60,000 and only have $5,750 out of your pocket. If they were to put 2.93 Billion of cash into the futures market they would control over 30 Billion of oil which would then make there fund a 15x ETF. This is similar to how ProShares and Direxion are able to produce their 2x and 3x portfolios with a small investment using leverage.

Hopefully this makes sense.

Sorry Kevin. You’re absolutely right. It’s becoming painfully clear that I was having some kind of brain malfunction yesterday. This wasn’t the only relatively simple thing I managed to completely goof up. 🙁

Ducati, Kevin, John, thank you all, really good comments …

per Ducati’s comments in his blog:

..The spread between March 2009 and June 2009 contracts grew to as much as 14%, which meant that by accepting March delivery and selling June it was possible to lock in 35% returns over only 90 days, minus the cost of oil storage…

Could the contango issue be a major explanation for the recent rally in Crude Oil and the Baltic Dry Index?

I mean, these rallies started pretty much at the time of roll-over of the March contract.

Should we expect a sell-off in both markets as the contango effect reverses?

Leave a Reply

Your email address will not be published. Required fields are marked *