Lynne Kiesling
In a full-information equilibrium in a liquid market, the forward price should exactly equal the spot price plus the carrying cost (storage cost plus foregone interest)[NOTE: I updated this to change the typo that the first comment caught, good catch!]. Thus you can explain the current increase in the crude spot price as driven by an increase in current quantity demanded because of the arbitrage opportunity between the spot price and the expectation of the price of the forward at contract termination, assuming constant carrying costs.
Isn’t economics cool? I’ve been marveling over this all morning. But the missing piece here that I’d like to understand better is the market fundamentals that are driving the expectations of greater relative scarcity later in 2005. What’s driving the high futures prices? Expectations of lower quantity supplied, higher quantity demanded, or both?
You got it backwards. In a full-info equilibrium market, the forward price should equal the spot price plus carrying costs, not the reverse. So in a well behaved futures/forward market for a storable product such as oil, contango–where prices further out are higher than spot or nearby prices–should be normal. If, and this is a big if, the oil market were a well behaved market, it would normally be in contango. Problem is historically oil and oil products have not been well behaved, and in fact the forward prices have been in “backwardation,” i.e., lower than spot, or the deferred contract lower than the nearby. This is usually ascribed to a “convenience yield” or some such, but it is definitely a departure from the pure cost-of-carry, arbitrage pricing model that textbooks say should apply to storable commodities. So today’s contango market for crude, while it seems “normal,” is relatively unusual for the oil market. Such oddities make economics more interesting, indeed!
Oooh, thanks for catching that. That’s what I get for a hurried post on the way to catch a plane.
Help me out a little here. After all, I’m just a DOUG…
What assumptions are you imbedding in your “full-information equilibrium” restriction? I would think that “full-information” means that everybody in the buying/selling market has the same information at the same time. But does that mean that they interpret the information exactly the same way? Full-information doesn’t mean perfect foreknowledge, does it? I would think that even in full-information, different agents would have different risk appetites, and might view uncertainties differently. And uncertainties must be present even in full-information, especially if there are possible constraints and/or thresholds in the “system” that produces and delivers the commodity. Future demand is uncertain, the system components are uncertain (explosions, hurricanes, terrorism), and the effect of constraints (storage capacity, deliverability) may be uncertain, if they are a threat to be binding within the bounds of all of the other uncertainties. This implies that the expected value of the future price may have a very wide distribution, and even under full-information I would think that different agents? perceptions of these risks might be all over the map.
Reflecting further, those different risk appetites would appear in the market as different bids and offers that would eventually resolve to a single price. But can that price be characterized as simply as you imply? What is not knowable is the expected value of the future price, since that is different things to different agents. So, perhaps, are you suggesting that a ?market expected value? can be implied by the current and forward prices, the expected value of the cost of storage, and some risk-adjusted measure of time-value of money? (Note that if storage and/or deliverability are such that they may or may not be binding in the period, then you really don?t know what the cost of storage is. It is uncertain, and among the market agents it will have a distribution of expectations.) Even so, while you can observe the price difference between two delivery dates, you can?t really know (even under full-information) the components that would establish the simple relationship you describe. After all, if these things could be known, we wouldn?t need a market would we? Isn?t this the knowledge problem?
If the market is expecting another bad hurricane season, which will disrupt Gulf of Mexico supply, wouldn’t that force up futures prices? Same issue if Bush strikes at Iran in 2007, for which I give even odds. Gulf (Persian Gulf, that is) supply would be disrupted, raising futures prices.
If the market is expecting another bad hurricane season, which will disrupt Gulf of Mexico supply, wouldn’t that force up futures prices? Same issue if Bush strikes at Iran in 2007, for which I give even odds. Gulf (Persian Gulf, that is) supply would be disrupted, raising futures prices.
Forward price should not exceed spot plus carry.
But it can be less. Al it takes is for the liquid
market to decide “a hamburger tomorrow is not worth as much as a hamburger today.”
You can use the carry to arbitrage-away a high
forward price. But you cannot carry backward, right?
Forward price should not exceed spot plus carry.
But it can be less. Al it takes is for the liquid
market to decide “a hamburger tomorrow is not worth as much as a hamburger today.”
You can use the carry to arbitrage-away a high
forward price. But you cannot carry backward, right?
I would think that at equilibrium in the absence of constraints, the spot price and the forward prices would have some dependable fundamentals-based spread. But if there are possible constraints in supply, transportation, storage capacity, etc, then I would think that the forward price could easily exceed spot plus carry. Of course, another way to look at that is that the carry cost is highly uncertain if carry capacity is possibly a binding constraint.
I would think that at equilibrium in the absence of constraints, the spot price and the forward prices would have some dependable fundamentals-based spread. But if there are possible constraints in supply, transportation, storage capacity, etc, then I would think that the forward price could easily exceed spot plus carry. Of course, another way to look at that is that the carry cost is highly uncertain if carry capacity is possibly a binding constraint.