Friday, July 14, 2006

Debunking the Geo-Political Premium in Oil Prices

I am currently seeking feedback on the following argument (meaning that it seems intuitively valid to me, but I am not convinced of its completeness or actual correctness—I’m looking for loopholes and problems. I would love to hear if you disagree with me and why):

It is commonly reported that a large portion of the price of a barrel of oil—especially a large portion of the recent run up in oil prices—is actually a speculative premium driven by fears of geo-political events. While possible future events such as war with Iran, major terrorist attack on oil infrastructure in Saudi Arabia, or escalation of violence in Nigeria could create a very real supply disruption that would justify such a premium, there is a limitation on how far and how long such speculative premium can elevate prices above the equilibrium point of current supply and demand. Oil is not subject to the same kind of wildly speculative forces that can drive, say, a bubble in technology stocks because, at the end of the day, an actual and fixed commodity is subject to delivery. While economics is not especially well adapted to mathematical proofs, what follows is an attempt at an “economic proof” that the risk-premium is in fact a negligible component of current oil prices:

- Oil producers are currently pumping oil out of the ground at the maximum possible rate—there is, for the sake of this proof, NO economically meaningful spare capacity for light, sweet crude oil

- When oil is pumped out of the ground, a producer has two economically viable choices: deliver the oil for sale to a consumer or store it to deliver for sale to a consumer at a later date. Regardless of who actually stores the oil (producer, middle man, storage of refined products, etc.), oil (and its derivative products) must be either consumed or stored.

- There is a limited global storage capacity for oil (and its derivatives), and storage costs are themselves a commodity with the price for storage increasing with demand over the short-term.

- The spot price of oil cannot incorporate any risk premium IF it there is no alternative option available to store that oil or to store any of the refined products of that oil. In a world without storage capacity—where oil must be consumed at the rate at which it is produced—concern about the availability of future supply cannot factor in to the price equilibrium between current supply and current demand.

- In a world WITH the ability to store oil (and associated refined products), the price of that storage (the equilibrium of demand for storage and supply of storage capacity) correlates directly with the price-premium in the market due to fear that future supply will decrease. Put otherwise, if fears of a geo-political supply disruption increase, then the incentive to store oil today to take advantage of the future higher price of oil also increases, creating increased demand for oil storage facilities, which is reflected in higher oil storage rates.

- Crude oil tanker fleets are a capital asset that are most efficiently utilized when they are in continuous operation at maximum capacity—meaning that if oil is to be stored, it makes the most economic sense to store it at the consuming nation, keeping your tanker fleet in optimal utilization.

- The United States is a major consuming nation of imported crude oil.

- National, strategic crude oil storage facilities (Such as the Strategic Petroleum Reserve or the rumored Saudi “tank farms”) are designed to address strategic concerns and are not responsive to market influences. Therefore, they do not themselves significantly influence the market dynamic of risk-premium and price of oil storage.

- There is very high elasticity of supply of oil storage facilities over the medium term (the term required to build an oil storage tank). If people are willing to pay a lot for storage of oil at existing facilities because there is a shortage, it is relatively easy to build more storage tanks. This high elasticity of supply means that, over time, a significant terror premium will not increase the price of oil storage, but will rather increase the demand equilibrium point—it will lead to the construction of more oil storage facilities, the supply and demand for oil storage will return to an equilibrium price roughly equal to what it was before the demand for oil storage jumped, and there will be more oil in storage.

- In other words, a risk-premium will drive increasing oil inventories. It doesn’t matter how tight supply is, if it is more profitable to sell oil for delivery at a future date than it is to sell it for present consumption (NOT merely sell it now), then oil inventories will increase as oil is stored for future delivery. It is actually this oil that is unavailable for current consumption because it is more profitable to store it that decreases de-facto supply and creates a higher price equilibrium—what is commonly called the “risk-premium” or “terror-premium.”

- An increasing risk premium will, then, lead to an increase in the amount of oil stored at ever-expanding storage facilities in consuming nations such as the United States. This means that if a risk-premium exists and has increased over the past few years, then this will cause a corresponding increase in oil inventories year-on-year in end-consumer nations such as the United States.

- This is not happening—in fact, inventories have remained relatively steady (perhaps declining slightly, and with seasonal variations) in the United States over the past few years. Therefore, there is no significant risk-premium in the current price of oil.

Who cares if there is or is not a “risk-premium”?? The risk-premium concept is, in my opinion, the convenient smoke-screen used to prevent discussion of the real issue: this is a supply and demand driven problem, and we are rapidly approaching or passing peak oil. Sure, geo-political events may make traders jumpy as they unfold, but it is my opinion that any price increase “on nerves” is quickly checked by the fundamental supply/demand picture. In this way, geo-political events act like a lubricant, but are not a prime mover themselves: they provide random inputs (and, psychologically, easily justifiable inputs in the minds of traders) to the markets, but a geo-political spike (or dip) will quickly bounce back to prior supply/demand equilibrium if it was not supported by some behind-the-scenes, supply/demand pressure in the corresponding direction.

12 comments:

LJR said...

Let's say I buy a hundred barrels of oil at $60 and have them placed in storage somewhere. I pay $1 per barrel per month for storage fees.

I put an ad in eBay saying I'm willing to auction my oil to the highest bidder.

Somebody looks at the headllines, gets freaked out and bids $80 a barrel. He believes oil will go to $100 a barrel in just a few months.

I take the $8000 and he takes possession of the oil and starts paying the storage costs.

As far as I can see, the oil in storage now includes a $20 terror premium without requiring any additional storage.

This is to refute your argument that the only way to make oil prices go up, given a steady production rate, is by building more storage.

I think you're missing the concept of a price being set at the margins. Oil speculators can bid up the existing supply of oil that is FOR SALE (a very small percentage of the total) just as housing speculators can bid up all housing prices by bidding up the price of existing and new homes that are FOR SALE.

Most oil is sold well in advance through futures contracts to the majors and remains off the market. The small puddle left over for trading can be subject to wild price swings.

Furthermore, we assume the mechanism used to price crude accurately reflects the "real" price. But that implies that if the price gets out of whack there will be an arbitrage to bring it back in line. In other words, if I believe the current price of $80 is too high I should be able to short it in such a way as to drive the price back down.

I don't think it's clear there is such a mechanism in the pricing of WTI or BC.

The entities who make out from the current scheme are the oil majors. They buy oil on fixed contract from National Oil Companies at much lower prices than spot. They refine the product and then price it at the pump based on spot. It's a wonderful way to collude since the figures they use are considered the "real" price of oil.

And this means the consumer pays for the "terror" premium caused by speculators driving up the price of oil at the margins.

It's very similar to the volatility of a stock being related to the "float." If a large company only has 10% of its shares available for purchase the result is often tremendous waves since the purchase or sale of relatively few shares can affect the price.

slx said...

To show convincingly that the risk premium and the scarcity premium on the price are two separate tendencies, maybe you could make a graph that shows the price fluctuations with the hidden rise subtract. The remaining graph should still show the geo-political premium when something happens, if you're correct.

Jeff Vail said...

The problem with the eBay $20 terror premium is:

-There will always be a few people who are grossly off in their prediction of future prices, but the efficient markets hypothesis weeds them out--market price represents the aggregate of all opinion about how much that barrel of oil is actually worth, so if you can buy them at $60, there will be just as many people selling at $40 (affraid of a bubble burstin) as there will people offering to purchase at $80.

- Most oil is actually not traded in futures contracts, but rather in forward contracts--the exchange trade in futures contracts IS actually the very "fringe" trade that, as you point out, sets the value for the next round of forward contracts.

- My argument doesn't really boil down to "the only way to make oil prices go up, given a steady production rate, is by building more storage." Rather, my argument is that rising prices, given a steady production range, require EITHER more oil being stored (resulting in more storage being built) OR an increase in marginal demand/decrease in marginal supply. The OR is what I think is actually happening, as suggested by the evidence that we are NOT seeing more oil being stored.

But, as I noted above, I could be thinking about this entirely incorrectly. The more I try to wrap my head around it the more difficulty I have--it is not an issue that is easily isolated. I think that you have correctly boiled this problem down to identifying the arbitrage play to defeat a geo-political premium that is not currently and directly impacting supply. What is the oil-risk carry? I don't know, but if one can find that--or demonstrate that it does not exist--then I think this problem may be solved either way.

Any takers?

Marc F said...

(1)
Why should the fact that oil is a commodity make it immune to irrational pricing by a market bubble. Stocks represent an ownership interest in a company that can be valued by the company's present/future earnings. Neverttheless, the price of tech stocks in the midst of the internet bubble bore no relationship to the earnings potential of the issuing companies. The speculators participating in the buble were able to price of stock shares to levels that were unjustifiable based on the fundamentals. The same can happen with oil. That is not to say that there is an oil bubble now. I don't think that there is, bU t I think there is no reason to think that there could not be one. Sometimes the markets don't price rationally.

(2) I was thinking about the airlines industry and whether researching the way in which various airline companies try to hedge against the rise in oil prices could be used to determine whether there is a geopolitical risk premium contributing substantially to the current rise in prices.

How to airlines hedge against a rise in the price of jet fuel? Do they purchase oil futures or stockpile jetfuel in storage facilities. My understanding is that they bet on the future price by entering into futures contracts with suppliers. Has anyone researched how the airline industry has hedged against what they believe will be a spike in oil prices in the short-term? Wouild this reveal anything interesing in the context of this debate?

LJR said...

Jeff,

I applaud your effort and your willingness to say you might be thinking about this issue the wrong way. My thoughts/counterpoints are ventured in a completely friendly way.

What I don't know is how the spot prices actually get set.

Here's a link that both of us might find helpful. I'm wading through it to see if it will help resolve my difficulty in understanding how oil is priced on the spot market.

www.eia.doe.gov/oiaf/servicerpt/derivative/chapter3.html

I, like you, think the issue of whether arbitrage can be used to cause reversion to the mean is at the heart of the debate.

My homey eBay example was intended only as an analogy - and probably not a very good one at that. I was simply pointing out that a risk premium for oil (or anything else) doesn't imply the need for more storage. The physical good just sits where it is while bits get moved around.

Oil is auctioned and the price is set by what bidders are willing to pay. I really don't see how it's that much different from housing stock, corn, grain or orange juice. It's just another commodity. Would you accept that extreme weather can cause grain and orange juice spot prices to fluctuate dramatically? If so, why should oil behave any differently? And why would a "terror" premium be any different for oil than a "hurricane" premium for Florida's orange juice crop?

The only reason I can see to make it different is that the spot oil price is somehow disconnected from the prices really being paid.

If prices are truly set at the margin then a crucial question is how sensitive is the market to purchases? If I had a billion dollars and invested it in spot oil all at once, how much would a barrel of oil go up?

I would buy all the spot oil available, starting at the lowest offered price and continuing to higher and higher offers until my money was gone.

The price for the next barrel of oil would be whatever remains on the offer ladder after I've bought my share.

Obviously this is a contrived scenario but I think it's informative.

The market has noticed my attempt to corner it and suddenly everybody and his brother has oil to sell since the new price has tripled from what it was when I started.

Suddenly all sorts of new offers come on the market at a lower price than that of the last barrel I bought. It's highly unlikely that my action would stimulate buyers vs: sellers unless the market figured I knew something they didn't know (a trick that has been used, by the way).

What occurs to me is basically what you said in your original piece. I've got to sell the oil to an end user at some point. In the meantime I will have to start paying storage fees for a lot oil and that keeps me from treating it like money in a bank account.

Let's assume I purchased the oil as a long term investment. My thinking is that, even at the ridiculous price I paid, the price will be much higher in three years. So much higher, in fact, that I can pay the storage charges an sell it three years hence still making a large profit.

This is where your assertion that more storage would be required starts ringing true. If I am holding a million barrels of oil off the market then I have to store it somewhere.

Let's leave the Warren Buffet scenario for a minute and consider my own lowly situation. I have $75 and decide to invest it in the USO ETF. I now own the right, in perpetuity if I choose, to sell a barrel of oil in the future at whatever the future spot price may be.

What are my storage fees for this barrel of oil? How does USO hedge my purchase? Does it have to pay a storage fee? Or does it simply buy options? If so, am I actually holding a barrel of oil off the market when I buy a share of USO? If I am not then how does that purchase reflect into the spot market? Somehow the cost of an option has to work its way through the chain and bid oil up.

I have no idea how that works.

Suppose Warren decides to make his billion buck purchase via USO. What happens next? Obviously the money in USO has to eventually find its way into the spot market and drive prices up.

Like you, the more I think about these issues the more confused I get. I used to think financial stuff was beneath me but I'm starting to understand just how complicated it can be.

Now my head hurts and I feel stupider than when I started.

Big Gav said...

While your reasoning seems sound enough, I think history shows that the idea of a "terror premium" isn't necessarily invalid.

Large price spikes have occurred in the past due to geopolitical tensions even though there was a massive over-supply of oil relative to demand.

So the current (albeit much longer lived) spike could empirically be considered the same as the spikes in the early 1970's, 80's and 90's.

The "supply shock" in 1973, for example, resulted in an enormous price spike - yet total world production grew 10% that year (just as it had in preceding years) - and there was a massive amount of cheap, easy to extract middle east oil available.

One Salient Oversight said...

My own research appears to confirm that an increasingly disparity exists between the amount of oil consumed and the amount of oil produced.

This suggests that the Peak is nearing, or has already passed.

See here for details.

Anonymous said...

This may help. Nervous speculation is only one factor in globally high oil premiums. Supply and demand for crude is also impacted after extraction & any storage by refining capacity, which, according to this episode of PBS's NOW (http://www.pbs.org/now/transcript/transcriptNOW145_full.html) is deliberately kept low by the major oil companies to ensure high prices. It is worth reading the transcript.

Dan Bednarz said...

Jeff,

Thanks for your argument. I'm in healthcare and mostly a socioligist, a little bit of policy analyst too.

I've been thinking about your argument in this way: what is going on is a widespread use of what Erving Goffman called "Cooling Out the Mark". This refers to masking really bad "it'll-freak-you-out" news with palatable small scale pieces of bad news. BY the time the truth is known the "Mark" is more-or-less already prepared for it. This can go on either as conscoius manipulative strategy or as self-delusion, cooling oneself out too.

And by the time the truth is know it's very late in the game, typically too late for the mark to do much more than cry or get angry.

CME said...

Jeff,
Thanks for the interesting argument. I'll take a stab at finding a weak point in it:

You said (accurately, I believe) that the Strategic Petroleum Reserve does not respond to market demand. You also say, "There is very high elasticity of supply of oil storage facilities over the medium term (the term required to build an oil storage tank)."

This strikes me as a potential weak point in the argument.

Are new oil storage facilties easy to site, permit, and build? What are some examples?

How much does it cost to store oil in a new storage facility? This would include the pro-rated cost of the facility itself, plus the costs (and any losses) involved in filling it and then withdrawing the oil later. Any "terror premium" would need to be higher than this cost, in order for increasing storage facilities to be a rational outcome.

Here's a counter-hypothesis: There is a signficant terror premium now built into the price of oil, but market actors believe that something will happen in the medium term (the time it takes to build oil storage facilities, or the time it takes to build them and recoup their costs through storage fees) to reduce that terror premium later. This could be discovery and exploitation of large, new oil fields in peaceful parts of the world, or the outbreak of world peace, or a world economic depression. Therefore, new storage capacity is not being built.

I'm taking your word for it that new storage capacity is not being built, by the way. I'm just an interested layperson with a little training in economics.

I look forward to your response.

Amin said...

I believe that major oil producers (OPEC) are not selling oil in the forward market. They are only selling Oil in the spot market. A couple of reasons for this. Most importantly, the type of oil they produce is not deliverable against the major oil contracts in the world (WTI or Brent). Unless, we have a situation like we have in the bonds where the concept is used "Cheapest to Deliver" where a range of bonds may be delivered, we will have a big vaccuum due the absence of major oil producers.

Secondly, the oil consumers, those that were previously not too concerned about oil prices have now an active portfolio of forward positions. Asian countries with new found wealth have figured out that its better to hold oil then to hold foreign currency reserves (which seem ominous to foreign government clamouring for them to revalue their currencies). We find more and more oil storage facilities being built in these countries. Unfortunately, the world only hears about oil inventory in the USA and not too much is known about other countries.

I think there is a risk premium in the oil prices and that is the risk that financial assets are losing their characteristic as the store of value. Same way as as soon as the USD lost its charm back in the 20th century and investors started swapping their USD for Gold at the Federal Reserve. Now its time for investors to swap all types of commodities against USD.

Amin said...

Derivatives in all forms be it futures or forwards or options are in essence cash and carry trade. That is if there is no producer that is selling forwards, then the financial intermediary will step in and create a forward price by simply buying the crude in the spot market, store, insure and finance it till the future date.

As companies try to reduce the volatility of their purchase price buy more and more of their oil through derivatives, the future demand actually starts appearing as spot demand.