In yesterday’s post I said that Chesapeake’s management was speculating on natural gas and oil prices. But Chesapeake claims that it is a hedger. Speculating and hedging are different things, so is Chesapeake a hedger or a speculator?
In representations to regulators, Chesapeake’s Vice President for Finance and Assistant Treasurer, Elliot Chambers, has stated categorically that “we never speculate.” Is that true? Is his definition of a speculator the same as mine?
Chesapeake is a hedger. It uses exchange traded futures and options, OTC swaps, and a specialized financing vehicle called Volumetric Production Payments, among other things, to mitigate the price risk on its production and “predict with greater certainty the effective prices we will receive for our hedged production.” (10K for FY2011 p. 72)
But Chesapeake is also a speculator. The company is straightforward in its SEC filings that it tries to profit off of price swings: “We intend to use this volatility to our benefit by taking advantage of prices when they reach levels that management believes are either unsustainable for the long term or provide unusually high rates of return on our invested capital.” (10K for FY2011 p. 6) “Depending on changes in natural gas and oil futures markets and management’s view of underlying natural gas and oil supply and demand trends, we may increase or decrease our current derivative positions.” (10K for FY2011 p. 59) “Our general strategy for attempting to mitigate exposure to adverse natural gas and oil price changes is to hedge into strengthening natural gas and oil futures markets when prices allow us to generate higher cash margins and when we view prices to be in the upper range of our predicted future price range. Information we consider in forming an opinion about future prices includes general economic conditions, industrial output levels and expectations, producer breakeven cost structures, liquefied natural gas trends, natural gas and oil storage inventory levels, industry decline rates for base production and weather trends.” (10K for FY2011 p. 87) As I related in yesterday’s post, Chesapeake’s decision last fall to remove its natural gas hedge was based on its prediction that prices were temporarily low and would recover, allowing it to replace the hedges and capturing a profit on the short-run volatility. This is all speculation.
A company can be both a hedger and a speculator.
When a company thinks it can forecast a swing in commodity prices, and it buys or sells derivative contracts to profit from that, it is speculating, pure and simple. The fact that the company doing this also happens to have a natural long position due to its production of the commodity doesn’t change that fact. And when the company has a natural long position with a partial hedge in place, but then adjusts that hedge in order to take advantage of forecasted price swings, its portfolio is best described as a combination of a hedge and a speculation. Stock portfolio managers are familiar with this decomposition. In a famous 1973 paper, Fischer Black and Jack Treynor showed how to integrate information on stock alphas into a larger portfolio. The portfolio can be decomposed into the passive, well diversified portfolio one would hold absent any information on stock alphas, and the active portfolio consisting of the variations from the passive portfolio that represent the bets on the information about stock alphas. Cheseapeake’s derivative strategy is much the same: one part a passive, plain vanilla hedge of its natural long position in production of natural gas, and one part an active, speculative bet made up of deviation from the plain vanilla hedge.
Like many companies, Chesapeake doesn’t like to think of itself as a speculator. It prefers to talk about optimizing its hedge. When its Assistant Treasurer insists the company cannot be a speculator, he has a clear and simple criterion: “we are strictly prohibited from hedging more than our estimated underlying production for any given future month, meaning we cannot and do not speculate.”
Here’s a little thought experiment that helps to expose the inadequacy of that criterion. Suppose we have two companies: one is a natural gas producer that does only does plain vanilla hedging with no attempt to time price movements, while the other is a small hedge fund that only does speculative trades in natural gas derivatives targeted exclusively to profiting from short-term price movements. Now, imagine that the natural gas producer buys the hedge fund, but keeps the two operations as separate profit centers. However, to minimize transaction costs, it centralizes all trade execution and maintains one aggregate portfolio. Its derivative position at any time equals the sum of the position its production arm would have as a hedger and the position its hedge fund arm would have as a speculator. Let’s also suppose that the hedge fund’s long positions happen to never be larger than the producer’s short hedge, and the hedge fund’s short positions happen to never be more than the producer’s net exposure. The combined company’s portfolio always satisfies Chesapeake’s definition of not speculating. But clearly there is speculation going on. When the producing company bought the hedge fund, the hedge fund continued doing the very same operations it was always doing. The national economy didn’t suddenly have fewer speculators just because of the corporate organization under which the speculation was organized.
Chesapeake is a hedger. But it’s a speculator, too. And it’s always useful to call a spade a spade. There may be nothing wrong with it being a speculator. Certainly it would be hard to make a public case for any legal prohibition against the speculation. It’s a private company, and if that’s what the shareholders want it doing, then so be it. But in order for the Board and shareholders to effectively evaluate management, it is necessary to think these things through clearly. And part of that is effective measurement of performance. The evaluation of a hedge is quite different from the evaluation of a speculation. Smushing the two together and trying to evaluate them in combination cannot be effective.