Wednesday, November 01, 2006

An update on the energy stocks vs futures arbitrage trade

I argued before in the beginning of October ("An arbitrage trade between energy stocks and futures") that energy stocks are overvalued relative to energy futures. At that time, a portfolio of long 1 front month QM (crude oil Emini future contract) and short 640 shares of XLE (energy stocks ETF) has a value of -$2,584. Where is it now? As of the close of October 31, December QM is at $58.725, while XLE is at $55.73 a share. The portfolio is now at -$6,305 (the multiplier for QM is 500). The spread has clearly widened: it is now at a 3-year low.

We are now faced with the usual arbitrage trader's quandary. Is this an unprecendented profit opportunity to double up on this trade, or was this a colossal blunder on my part? I came across this New York Times article about the earnings reports from Exxon and Shell that gave me some comfort. While both energy companies posted huge profits, the article quoted Fadel Gheit, a senior energy analyst at Oppenheimer & Company, that for the fourth quarter, "“the question is not if earnings will decline, the question is by how much.” According to the article, analysts say that for every dollar the price of a barrel of crude oil drops, Exxon forgoes $500 million in profit.

So yes, with my fingers crossed, I am still waiting for the day when this spread closes up.


Eric said...

What makes you think there should be a significant correlation between the front month oil contract and the price of an energy equity?

Ernie Chan said...

The reasons are 1) the last 3 years of data show highly statistically significant "cointegration" between energy stocks and front-month contract. Go to my original article for the plot of the spread and you can see for yourself. 2) Energy companies asset value is in large part determined by the value of the commodity they own -- crude oil. 3) Cashflow of energy companies are also determined in large part by the price of crude oil (see the NYT article I quoted).

Jim said...

Ernie, interesting article. I can think of a few issues with only 3 years of data (I understand limitations of older data).
1)longer time frames allow management to add value to the commodity: buying property when the price is low, etc. XOM over 10-20 years has advanced nicely vs cash oil mkt thru superior capital management.
2)potential hedging. Not in widespread use, but innovative nat gas companies are hedging more to lock in the strip price and protect earnings. This will throw off your model as their earnings and price performance may resist a direct relationship with underlying cash volatility.
3)bull market. Intangible, but energy awareness has been high for the past 3 years. Trades in E&P equities are crowded and popular. This has not always been the case. Perhaps this is an argument that equities are overvalued as you say.
4)peak oil is here. (?) If this is true, the relationship of recent past is broken as energy in the ground is worth much more than just earnings. Access to drilling is scarce, and deserves a premium. Oil sands trade at a much higher valuation vs Majors for this reason.

Great work. Ever do anything with options? I have a few questions about anomalies that I have been unable to find answers too.

Ernie Chan said...

Jim: Great insights. I completely agree with 1) -- there may be a long-term out-performance of energy companies over energy commodity prices. But since my trading horizon is over a few months only, I hope to neglect that. 2) A short-term breakdown of the relationship due to hedging may present a short-term arbitrage opportunity as this relationship will be reestablished again as their futures or options hedge expire. 3) Very true: XLE has a positive beta to general stock index, not just a positive beta to commodities price. So if and when the stock index takes a dive, it is time to exit our arbitrage trade profitably. 4) Over the longer term, if peak oil comes into play and we have to get oil from oil sand, should we not expect higher oil prices over all? That would then drive the reversion of the current wide spread back to zero also.

I would wait for another quarter to see how oil companies fare under the recent lower-oil-price regime and see if the long-term cointegration relationship has broken down or just temporary suffering a huge distortion.

I haven't done anything with options, but am happy to chat about them. Email me if interested! Thanks a bunch for the great comments.

Eric said...

I was not taking exception to the cointegration of XLE and oil prices. I was taking exception to the use of the front month contract as the most effective trading/hedging vehicle. The value of an oil producer's stock is the PV of a long dated string of cash flows. It is certainly true that these cash flows are determined by crude oil prices, but not the prompt contract. If a company has a 10 year reserve life, why should the value of the prompt crude contract be determinative of the stock's value. I don't understand why you would subject your strategy to the volatility inherent in the prompt contract when it is easily avoidable by moving out the curve. Are you familiar with the stack and roll blow ups ala MG in the early 90's. What you are discussing is much the same concept.

I think there is a great deal of validity in looking for disconnects between equity values and commodity prices. However, trading a prompt commodity contract versus a long lived asset is not theoretically justified.

On another note, when looking at your data what do you do with the roll. For example we are rolling up over $2 in crude when January becomes prompt on monday.

Ernie Chan said...

Eric: Very thought-provoking points that you have made. You are absolutely right to say that the prompt contract is not the best hedging vehicle for energy companies stocks because it doesn't reflect the PV of the companies. You are also very right to say that the price of the prompt contract is especially volatile. But my interest here is not hedging, it is arbitrage trading: taking advantage of the temporary mispricing. If the mispricing is large, we have a larger opportunity for profit, volatility notwithstanding.

To avoid possible confusion and the complications of futures and rollovers, I'd like to construct this analysis simply as an arbitrage between the oil fund USO and XLE. Or if you like, between spot oil prices and XLE. In this case, would you agree that the current oil companies valuation is based on the expectation that future oil prices are higher than the current oil price? And if oil prices, hypothetically speaking, remain at the current level for another 12 months, would you not agree that people will start to revise downward the valuation for these companies? If that is true, than the spread will start to revert (converge) to get closer to zero.

Answering your question about the roll: I use my data provider's perpetual contract series. I have not checked whether they adjust the previous front contract prices upward based on a fixed dollar adjustment (in which case a P/L calculation would be correct) or on a percentage adjustment (in which case a return calculation would be correct.) I believe a fixed dollar adjustment would be more correct for cointegration analysis. However, since the conclusions of my analysis is the same even if we use USO as the instrument, I gloss over this point.

Thanks for the comments!

Jim said...

Two follow up observations:
1-have you tried the OIH instead of XLE? Drilling is more cyclical than pure E&P and there could be a more 'obvious' trend to trade against.
1-Instead of a cash or front month oil have you considered using the strip of 12 month fwd contracts? In the industry this is used commonly for price assumptions instead of a single month. It would be substantially less volatile and not subject to single month distortions -amaranth, hurricanes, etc. You cannot trade the strip in financial markets (but available as a swap to institutions), so this may only serve as an entry point signal and not a nice paired hedge. Jim

Ernie Chan said...

Jim: I personally have not tried OIH, but my friend Yaser Anwar over at did suggest that to me. Its spread with oil prices does seem to fluctuate less than XLE. I will post a study of this next week.
About the strip: it will also certainly reduce the volatility in the spread. But as I mentioned to my reader Eric above, I am not so concerned about volatility -- I am only concerned whether the spread will eventually revert. On the other hand, if I do want to trade this strip, I can just buy/short a portfolio of the 12 fwd contracts, right? There is no requirment that this trading strategy be implemented with just 2 instruments.

Thanks for the suggestions!

Anonymous said...

Not sure if anybody looks at comments to a 3 months old blog, but just in case...

The problem with using a 12 month futures strip as your commodity hedge is trade size. In your example you are looking at 1 mini futures versus 640 shares of XLE. Obviously the ratio will change slightly but using the same ratio and assuming you have to use full size contracts you need to hedge over 15k shares against your 12k bbls of oil. I once spent a lot of time researching this type of arbitrage but in the end the equity trade size put me off.

You mentioned possibly using the USO ETF instead of the prompt month mini contract. I think that would expose you to the same roll risk, except it would all be handled for you. (There's lots and lots of discussion on the effect that the huge contango is having on USO if you haven't seen it.)

A potential solution to both trade size and roll risk is oil ETF USL. Unlike USO it tries to mirror the 12 month strip price. You still have roll risk, but it's now 1/12th of the M1/M13 roll each month rather than the entire M1/M2 spread. With the prompt M1/M2 roll around -1.50 and the M1/M13 roll around -8.7 your effective roll cost is reduced by half to approximately 0.72.

I've never traded USL, but I have heard it is a lot less liquid than USO so execution costs may be a problem.

Ernie Chan said...

USL sounds like a good idea. For small accounts and long holding periods (as anticipated for this mean-reversal strategy), hopefully liquidity is not a big issue.