Financial Intelligence: How Arbitrage Forensics Provide Insight into Saudi Knowledge
Posted by Nate Hagens on March 29, 2007 - 10:24am
Topic: Economics/Finance
Tags: arbitrage, backwardation, contango, futures, iran, oil prices [list all tags]
The following is from guest contributor Jeff Vail. Jeff is an intelligence analyst focusing on energy and infrastructure-related issues. He is a graduate of the US Air Force Academy and a former USAF Intelligence Officer. Jeff previously wrote on the The Oil Drum about the increasing violence in Nigeria.
He discusses an interesting phenomenon with respect to energy futures prices, that long dated futures are limited in how much they can go up (but not down) based on arbitrage principles. Because the price of distant oil futures quickly rise alongside spot market prices when spot markets are moved by short-term events, we can infer that major producers such as Saudi Arabia think that the future price of oil will be much higher than the price at which distant futures are currently trading. This provides further support to the theory that they don’t believe their own statements on their future production or on the future price band for crude oil. Jeff's post is under the fold.
NYMEX Futures Strip for Light Sweet Crude 3/29/07 (Click to Enlarge)
Financial Intelligence: How Arbitrage Forensics Provide Insight into Saudi Knowledge
There has been quite a bit written recently about what is “really” happening to Saudi oil production, as opposed to what the Saudis are telling us. Comparison of rig-count to oil production levels, or analysis of published seismic readings of water encroachment on a reservoir provide good insight, as Stuart Staniford has shown in several, recent, articles. This article is to suggest that additional, complementary information about Saudi oil production can be gleaned through forensic analysis of the related financial markets. Just as reservoir analysis is quite complex, this financial analysis will be fairly challenging—please bear with me, I will do my best to explain the financial principles at work in this analysis (and please don’t be offended if you already understand arbitrage!).
EXECUTIVE SUMMARY: Because the price of distant oil futures quickly rise alongside spot market prices when spot markets are moved by short-term events, we can infer that major producers such as Saudi Arabia think that the future price of oil will be much higher than the price at which distant futures are currently trading. This provides further support to the theory that they don’t believe their own statements on their future production or on the future price band for crude oil.
One of the basics principles of finance is the “Law of One Price.” This states that, for any fungible and definable good (such as oil), all future prices flow from the spot price. Say, for example, that the spot price of NYMEX Sweet Light Crude is $60/barrel. The Law of One Price says that the future price of that same barrel of oil in one year is not an independent price, but rather is dependent on the spot price: it is $60/barrel + “risk-free” cost of money over 1 year + cost of carry (storage) for one year. If the future contract is trading at higher than this predetermined cap, then arbitrageurs can make risk-free profit—they sell the future contract, buy the spot oil, and store it until the future contract matures, collecting the excess as profit. This represents an upper limit, an “arbitrage cap”—the price of oil for delivery one year from today cannot exceed a set amount above the spot price, or arbitrageurs will act to correct this price distortion. It should be noted that arbitrage price cap can actually be lower than the spot price—this happens when the cost of money is negative over the time period (either due to general deflation or specific yield-curve environments).
In contrast, there is no limit to how much lower the future price of a good can be from the present price because arbitrage cannot act to correct such a disparity. If oil is trading at $60/barrel today, there is no reason why a future contract for delivery in one year can’t be trading at $30/barrel. A trillion barrel oil field could be discovered in upstate New York—not very likely, but we just don’t know. The point is, it is theoretically possible for traders to find some good reason to price oil much lower one year from now than today’s spot price. Arbitrage has no mechanism to correct this disparity—you can’t buy a future contract to cover a sale of oil today.
Returning to the upper limitation on future pricing: what happens when future prices push up against this arbitrage cap? Arbitrage acts to bring the prices in line with each other. But in doing so, does the future price come down, or does the spot price go up? Let’s look at the mechanics. An arbitrageur looking to exploit a future price that is too high relative to the spot price must buy oil on the spot market and sell oil on the future market—this increases current demand (relative to current supply) and increases future supply (relative to future demand). As a result, both the future price comes down, and the spot price goes up. However, the volume on the future market is invariably much lower than the volume on the spot market, so the arbitrage trades bring the future price down much more than the spot price goes up.
As we’ve discussed, the spot price of oil cannot “push up” the future price via arbitrage. The spot price does set the psychological expectations of traders (as humans tend to extrapolate the present when predicting the future), but there is no arbitrage mechanism that forces short-term, supply-based price increases (such as the shut-in production following Hurricane Katrina) to push up prices in the distant future. Why, then, do actual price changes seem to contradict these accepted principles? Now things get interesting…
When there are short-term, supply-driven increases in the spot price of oil—such as the recent increase that can be partially attributed to the capture of British sailors by Iran (see chart below)—this price increase should not also increase the price of a future nearly four years out. And yet it did…the price of the December 2012 contract increased right alongside the spot price over the past few days. Why???
Because of the limits imposed through arbitrage, the price of oil (based on the future contract price) in 2010 can’t greatly exceed today’s spot price of oil. The future price of oil can, however, be significantly lower than the spot price—and when short term events push up the spot price significantly, this might be normal occurrence. After all, an incident in the Persian Gulf, or a hurricane in the Gulf of Mexico have little bearing on the supply/demand picture for crude oil five years from now. But in the past few years, as soon as short-term spot price movements create space under the arbitrage imposed future price cap, the future price seizes the opportunity to move upwards. The explanation for this is that the future supply/demand equilibrium is at a price significantly higher than the future is actually trading, because that future price is constrained by the arbitrage cap. Whenever space is made available under the arbitrage cap, such as by current events, the future price will quickly rise to fill that space.
Figure 1: Illustration of distant future market closely following short-term spot market price movement to fill available space under the arbitrage cap.(click to enlarge)
If the future price of oil were significantly below the arbitrage price cap, then this would be highly significant to the Peak Oil debate—it would represent the market intentions of major producers who, with full access to their internal data, believed that the future price would not be significantly higher than the present price. Conversely, because there appears to be zero space available under the arbitrage cap—even when events temporarily increase the spot price—we can discern that these major producrs believe that the future price will be significantly higher. In reality, this calculus applies primarily to only one producer—as Saudi Arabia is the worlds largest exporter, and as most future hopes for oil supply increase seem predicated on their claims, then they are the only market player whose future production has a high likelihood of setting future prices.
If Saudi ARAMCO actually believed that it would be producing 12+ million barrels of oil a day in 2012, and that prices would still be trading in a band of roughly $60/barrel, then it would have significant motivation to sell oil futures for delivery in 2012. Why? If it can get $68 today for oil to be delivered in 2012, and it honestly expects oil to be trading in the $60s in 2010, it would be crazy not to sell the future today and invest the money in a bond or other financial instrument. Even at a 3% rate of return above inflation, selling a Dec. 2012 contract today for $68/barrel is the same as selling that same oil in Dec. 2010 for $81/barrel (in 2007 dollars). If Saudi ARAMCO did believe what it publicly states it believes, then it would be happy to sell oil for delivery in 2012 at $68/barrel (today’s future price). Or at $67/barrel. Or at $66/barrel—this motivation to sell even at prices below where the future is currently trading would bring the future price down—it wouldn’t jump to fill space created under that arbitrage cap by short-term influences. Because this is not what is happening—in fact, because the opposite is happening and future prices rise immediately to fill any available room under the arbitrage cap—we gain a very valuable insight into the inner thinking of the Saudis. What insight? Because the future price jumps to fill space created under the arbitrage cap, we can infer with high confidence that the Saudis don’t believe their own rhetoric. They don’t believe that they’ll actually be producing 12+ million barrels of oil per day in 2010, they don’t believe that the price of oil will be less than $80 in 2010. If they believed anything close to this, they would happily sell 2012 futures now for $60/barrel or less—far less than the current price—and this would bring down market prices. This is, in my opinion, highly significant because their access to their own data puts them in the best position to make assessments of the future of global oil markets. If actions speak louder than words, then this is the real Saudi press release: “We’ve peaked… ”
- "If the future contract is trading at higher than this predetermined cap, then arbitrageurs can make risk-free profit—they sell the future contract, buy the spot oil, and store it until the future contract matures, collecting the excess as profit."
Try booking 12 month storage at Cushing from the storage owners, they will extract the full rent the market pays in the forward contango and you will end up losing money. There is no such thing as a risk free profit except in text books.
- "If it can get $68 today for oil to be delivered in 2012, and it honestly expects oil to be trading in the $60s in 2010, it would be crazy not to sell the future today and invest the money in a bond or other financial instrument. Even at a 3% rate of return above inflation, selling a Dec. 2012 contract today for $68/barrel is the same as selling that same oil in Dec. 2010 for $81/barrel (in 2007 dollars). "
The seller of 2012 $68 futures contract will only get paid the money on delivery of the product in 2012. Thus the NPV (at 3% bond rates) is $55, not $81 (haven't checked the math, but going by the example given)
In simple terms, imagine agreeing to sell your house to someone else for $X, the sale to go through in 2012 and money to change hands then. To monetise this, you have to take the contract to the bank who will then discount the value by the interest rate applicable and the risk of the eventual buyer defaulting at some point between now and 2012.
The bank is going to discount by a rate higher than the average 5 year interest rate and then give you a credit haircut on top of that, and a risk haircut if your house is in Iran, Angola, Saudi, etc, for example....
In other words, though futures prices are nominally higher, they are lower than spot prices in terms of net present value.
Bunyonyead-
Two good points that both represent barriers to entry to the world of arbitrage for small-timers. But they don't apply to producers and institutional players--the ones whos behavior are dictated by these dynamics.
If I want to go book storage for a thousand barrels of oil, I won't get a great price. Major players, on the other hand, can demand efficient pricing for oil--because they have the leverage to own their own storage facility if industry providers aren't providing it at a competitive price.
Likewise, if you sell one 2012 future, not only do you not get the money now, but you have to actually put up money to cover your liability. Major producers, however, are treated differently. They get a significant chunk of the money today if they operate on the NYMEX, and they can do even better if they enter into forward contracts that track NYMEX prices.
Jeff,
Agree in theory. However.....
Building new tanks may be more difficult in practice than in theory, planning permission etc. You still have to pay to access pipelines, etc. Anyone building new tanks is doing it on the Gulf Coast where the real market for crude is, not in Cushing.
Major producers sell forward swaps to banks (Goldman, Morgan Stanley, etc). The banks discount them for interest rates, credit risk, location risk (resource nationalism etc), war risk, etc, etc, etc. Ultimately they get less than spot price on this...
I am not aware of any major producer selling futures strips on NYMEX, margins are too high and liquidity is too low. The banks do all this business and there is constant flow of forward producer selling in the swaps market.
Addendum: a LOT of forward producer business has been done in "costless collars" where producers buy a put option below the market and sell a call option above the market (eg $50 and $80). They are therefore guaranteed a minimum of $50 and a maximum of $80 and take floating price vs NYMEX for anything in between. There has been so much of this business in the last few years that put options are priced relatively higher than call options, known as an inverse risk reversal skew. Crazy, but true, and a reflection of real pricing flows in the market...
For those who may not be familiar with the term "risk reversal", it commonly refers to the difference in option implied volatility between the .25 delta calls & puts.
I first encountered these in the FX markets back before the Euro came into existence. "Back in the day", large volumes of risk reversals traded on Dollar-Mark, Dollar-Paris, etc.
Correctly modeling the skew is crucial to more than just the pricing of risk reversals, but to screwball derivatives of all types (e.g., single- & double-barrier knockouts & knockins, in-the-money payoffs raised to a power and then capped, etc.).
Trying to translate using my other post. I thin they are trading out to their strong probability positions.
Assuming I'm right and hedges are use to setup big win probability distribution and limit losses. At the end of the day I cant see how positions don't translate into probability maps of gains and losses with various crystal balls. This means even slight contango in the market snapshot is a big deal.
"They get a significant chunk of the money today"
Are you sure about this? My understanding is that commercials get a lower margin than specs but that it's still a margin. Your comment seems to refer to forwards (off exchange) rather than futures.
forwards are not settled until the positions are closed either. Low credit players might have to put up an L/C to cover the risk, but no money changes hand. Better risk players will have margining agreements with each other...
wrong
all players on the NYMEX have to post margin. It's less for the big boys with good credit, but there are no freebies. And no one collects on open positions until they are closed.
No one pays on forwards until they are closed either. margining takes place unless the entity is given open credit. No doubt Saudi would get open credit from Morgan, but no money flows would take place until settlement.
"buyer defaulting"
The exchange prevents this from happening in futures contracts by acting as the intermediary party -- and keeping margins current by daily settlement.
I also don't think it is correct to use the risk free rate to discount Saudi's cash flows.
The discount rate should be Saudi's cost of capital, or the return they would get on the next best investment with a similar risk profile. Storing a volatile commodity is hardly risk free, even though Saudi may have special knowledge and ability to influence the price. I would be an interesting exercise to try to calculate an appropriate cost of capital for this investment. There are an awful lot of factors involved. I would guess a reasonable estimate would be more like 7-8%.
the debate is meaningless. The saudis don't hedge. Neither does Exxon or any of the other majors on their forward crude production. The sellers down the curve are smaller players that have high cost production and banks eager to make sure their loans get paid.
Excellent article...I have been wondering when there would be a guest post on the signals from the forward oil curve. The only thing that has stopped me from sinking $50,000 into long dated contracts (or better yet, options on those said contracts) has been the fear of the margin call on a short term spike down.
The curve is not in perfect contango, so there is still some arb. cap beyond 2009....
The floor was back at the beginning of the year and even now it can pop down to $60, but with spring/summer coming and the market moving down gradually, shit hitting the fan won't commence until next year. This year will simply be a painful pin prick.
There's another thing to add to your comment. As SA refines more of it's own oil, less is sellable on the exchanges. This could have a weird effect. lately WTI has trailed Brent by a pretty wide margin, as much as almost $4 for a brief time. This is VERY significant.
As more refineries open in Asia and Close in America, the US petroleum reserve and WTI oil have less of an effect on price. Why would someone buy $68 light sweet in Beaumont for delivery in 2009 sitting next to a trillion barrels that can be fed to a dwindling number of refineries when they can pay $47 for the Saudi basket deliverable to indonesian refineries.
People seem to forget, benchmarks are only good if they reflect real supply demand. That's why the Omani benchmark isn't used any more. This is why the crack spread is so wide. Spot price is a supply chain phenominon, as Platts loves to remind us.
Excellent point. I have made a number of posts recently explaing why WTI is a poor benchmark for crude pricing. Cushing is landlocked and the pipelines servicing it are mostly owned by the refineries who also happen to own the storage tanks.....
Interestingly though, TEPPCO is allegedly discussing reversing the flow on the Seaway pipeline (currently flowing 280-350 kbpd from Freeport TX to Cushing) and will announce today the completion of the Takeaway pipeline from Midland TX to New Mexico (which will also reduce flows to Cushing).
Yes, it is significant, but what does it really mean? My take is that it reflects the premium for tanker delivery as opposed to Cushing delivery. That Asian refiner is not exactly at the end of the NA pipeline system.
it reflects a heavily contangoed market. Nobody really needs prompt wti and storage is full. The fear factor is boosting later delivery and more so in Europe as Iranian crude is actually refined there in a significant amount.
what????
Refineries may have closed in the USA, but we are refining a record amount of crude. The number of crude units is not meaningful. Their capacity is.
Brent/WTI is not really that inverse to normal. You are looking at prompt futures to prompt futures and there are no jet deliveries of crude oil to make that happen. When April WTI dropped to $56, May was $59. Brent was a only a little over May.
My guess is Brent is over WTI right now because European refiners actually use Iranian crude and are nervous about it. Also May/June is maintenance season in the North Sea. Its the period when the BSDs would try squeezes as production was at its min while demand was good.
Do we have any evidence that the Saudis have been active palyers in the futures market in the past? As the major swing producer, they would be foolish to do so-- nobody would trust them, or the markets, anymore. Sort of like the ultimate insider trading: they quietly go long, then announce a quota cut, or they go short, and then announce a production expansion.
I just don't think that we can use a market in which the Saudis do not participate to gauge their future actions.
Clear data on how Saudi insiders participate in the markets would require the very transparency that we don't have. Additionally, the ability to participate in the market is just as relevant as actual participation--it represents opportunity cost to all other actions.
The volumes on the NYMEX for the crude oil future contract don't appear to be sufficent for the major oil producing countries to effectively hedge. Although daily volumes have been exploding -- record contracts increased from 162,000 in 10/06 (each contract representing 1000 barrels of oil) to over 800,000 in 1/07. It seems to me that if Saudi Arabia wanted to lock in its future production of 7-10 million barrels per day it would need more than 160-800 million barrels per oil traded per day otherwise the price would be moved too much to the upside (although I'm not exactly sure how much more volume would be needed).
Although with low enough volumes on the futures exchanges KSA could (1) sign large volume long term contracts tied to the spot prices and (2) then manipulate the spot prices with very little effort to leverage those contracts. Still, given the volumes of liquids being moved every day and the other considerations KSA must have (eg, preservation of monarchy and the present tight connection between royal heads and royal bodies), I do not see them messing with the futures market that much.
With hundreds (?) of billions of barrels in the ground, and their entire economy dependent on oil exports, the net Saudi position is always long, never short.
For comparison, applying the same logic how do the future prices during the seventies fuel crisis look?
There were no futures then...NYMEX started trading crude in 1983 IIRC....
Heh, displaying my ignorance every day...
Hi there folks. I hope people will digg, reddit, and SU, etc., etc., this story for Jeff. It's a good one. Thanks!
Not to disagreewith what is said here, but the other looming variable is the actual value of the currency this future is denominated in. As was demonstrated quite siccinctly in the 70s, inflation - or rather more accurately the printing of money - can make a complete mockery of future prices. Given the US current foreign obligations, which a further rise in oil prices would only aggravate, I see no reason to expect that the correlation between today's relative value and that of 2012 would be any closer than it was then.
In Euro terms, the price of oil has not moved up nearly as much as in dollar terms. Or, put another way, the US dollar IN OIL TERMS has fallen much further. I'm more certain of the value of oil than I am of any currency including gold. How we manage to continue an historically growth based economy given the deflationary effect of a dwindling oil supply will be the condrum of the new century.
I'm more certain of the oil supply five years oout than the money supply.
That needed to be mentioned. it's hard to think in so many dimensions... crude grades, shipping, supply chokes, refinery locations and currencies. This is why markets work so well. We all learn how "smart" we really are.
I should have sUccinctly said conUNdrum - and that's a Freudian Canadian oout!
Fascinating article, thanks for sharing! I know almost nothing on this subject.
Many have pushed the argument (Michael Lynch, etc.) that the price increase since 2004 is mainly due to speculation (hedge funds, etc.). I would like to know your opinion on this? I would suspect that speculation would leave some kind of signature (e.g. number of non commercial contracts versus commercial contracts, etc.).
I think that speculation has had some effect, but that it is primarily an easy excuse that people can point to when they want to explain price increases without addressing supply issues.
Speculation is certainly an issue with financial instruments that aren't pinned to reality through delivery of tangible goods. But oil, as with some other commodities, is constrained by fixed supply and is traded in contracts that are subject to actual delivery. So while speculation can drive up the price of shares of Google, there is no "delivery" of some tangible "google" good at a fixed date--price is entirely theoretical. Oil prices are at least partially pinned to the reality of the supply and demand picture because it is traded in instruments that must be delivered at a definite date.
Bottom line: oil prices are the result of an equillibrium between supply and demand. I'm sure there is some degree of speculation (though it's difficult to tell who's speculating and who's hedging in some manner), but if speculation has driven prices up, this increases the incentive for producers who are capable to produce more today. The fact that production hasn't risen to burst this "speculative bubble" is, if anything, evidence that the bubble is not speculative at all, but rather the result of supply limitations. So I think that listening to talk of "geo-political premium" or "speculative premium" come from Lynch or CERA is evidence that they'll do anything to avoid talking about supply limitations. After all, academic economists will tell us that commodity prices should always go down!
the only place you can speculate is on the futures market. Kurt Wolff gets around this by using 6 year futures.
Right. Speculators can only influence price in the short-term and can only make money if they are right. If there was ample supply coming that would drop prices in the future, it is still possible that a crowd of stupid speculators would rive the price up, but when supply came on line, they would lose their shirts.
All speculators can do is to anticipate a price increase and buy in advance of it. This is true for oil, currencies, stock prices, real estate, etc.
Or that it wasn't a bubble. The speculative activity correctly anticipated the price increase and the market was right again.
it is also crucial to note, as you did, that it is not possible to divide the world into speculators and non-speculators. If a refinery hedges we say they are not speculating, but merely covering risk. but then if they decide not to hedge, are they then speculating? What about an investor with a portfoio that has oil exposure (say Korean equities), that enters into a hedge? There is a huge amount of overlap and one can "speculate" by doing nothing.
All of this is extremely interesting, and it sounds very scientific. Yet, looked at more closely, it sounds like maybe the discussions between acolytes of Tiresias (the Greek seer who ultimately led to the discovery and downfall of Oedipus). Are we looking at "real" facts, or bird entrails, or copulating snakes? And how is anyone really to tell the difference?
Seems to me that arbitrage is a giant casino, with the odds in favor of the house, and the players competing under significant and unequal financial handicaps -- but individually they win just often enough to keep them coming back for more. (The principal of random reinforcement.) My friends and acquaintances who gamble all say the same thing when they come back from Reno or Las Vegas-- "I had a great time, the food was cheap, and I broke even." Well, of course, those places wouldn't be there if everyone either broke even every time, or even more important, couldn't be made to believe he broke even.
What's the point? Well, there are a lot of variables that the oil arbitrageurs don't and can't know or control, and the whole thing can spin out of control (c.f. Long Term Capital Management, or the U.S. Savings and Loan collapse of the 1980's)-- in which case, the entire society is drawn in through taxation, giant financial manipulation or war to save the House.
An old financial partner of mine used to play poker at the Elks club every Thursday night, and he always made money. He played his cards straight, and he never drank. Some would say he never had any fun, and he wasn't playing by the social rules -- his world of total moderation is certainly antithetical to a world that demands novelty and risk.
Digg'd your article, Nate.
Question for all: how does a peakist get in on the futures market? can it be done with $50-100k? Anyone making some dollars on the backs of the blissful MSM?
I want to buy 16 PV panels to shine at my neighbors mcmansion this year...
Thanks,
Charlie
Its Jeff Vails article - i just formatted it.
You can buy one futures contract which controls 1,000 barrels of oil ($64,000 worth) for margin of about $4000. I wrote about how this works here.
Basically you can put up as little as $4000 or as much as full price, depending on how much leverage you want. The more leverage, the more risk that a downtrade (due to recession or some such) will eat away your margin. Its risky but you dont make high rewards without commensurate risk.
You can open a futures account with any number of brokerage firms - I use Man Financial but there are others.
And as an interesting but scary bit of oil futures trivia:
Kind of bizarre when looked at in that way.
Personally, I prefer options on futures to owning actual futures contracts. They are probably not the right choice if you're a large player who will be trading frequently, but I think they're preferable for the small, individual participant. Options have three primary advantages IMO: 1) you can't lose more than you pay for the option, 2) the price of entry can be lower, and 3) you have more flexibility because you can buy options near the money or far out, and the lower price of options well off the money make it possible to buy several options--this allows you to more effectively use options to hedge against personal exposure to energy prices.
At the moment, a call option (i.e. going long on oil) for the December 2010 futures contract at a strike price of $100/barrel is selling for $2200. This means that, for $2200, you make $1000 for every dollar over $102.20 per barrel of oil on November 15, 2010. IMO, this is an excellent hedge--it is probably best to not consider it an "investment." It can potentially hedge against loss in suburban home value, loss in 401k value, higher home heating and gasoline bills, etc. But as with any investment, don't take my word for it :)
I think everyone on this board expects oil to be WAAAY above 102 in 2010. So would it be smart to put 50k in to a future option contract now or wait a bit.?
If I'm hearing you right I could make a couple of million dollars iff oil is 200 in 2010, which I think likely.
Am I right????
Is it that easy?