January 18, 2006
Oil market jitters
The February oil futures price on NYMEX has jumped $8 a barrel in the last three weeks. It's useful to try to put this into a broader view of what's going on in the world oil market.
The immediate impetus for the recent oil price hikes seems to have been unfavorable news developments in three key global hot spots. Iraq's unsteady oil production has been even less steady than usual. For neighboring Iran, bettors at Tradesports are now putting a 40% probability on tension escalating into a military conflict. A disruption in Iranian oil shipments for whatever reason would likely prove to be a very significant economic event. And turmoil in Nigeria is particularly unsettling given the critical importance of that country's oil production in everybody's scenarios for the next few years. Iraq, Iran, and Nigeria together account for about a third of the increase in oil production from OPEC that Cambridge Energy Research Associates is counting on over the next five years.
But it's also important to try to maintain a longer term perspective. Oil prices have been on a broad and very impressive upswing over the last four years, though up until early 2004 this had only brought oil back up to where it had been in 2000. In accounting for the price surge in 2004, it seems clear that developments on the demand side rather than the supply side played the key role. Global oil production was up 3.3 million barrels per day in 2004 relative to 2003, the biggest annual increase of any year during the last quarter century. And production in the first six months of 2005 showed robust gains over the corresponding months in 2004.
I argued last August that one needs to amend this story to describe developments as 2005 progressed, as supply limitations rather than demand increases came to be the dominant story. Indeed, global production fell by 0.3 mbd in September and 1.1 mbd in October compared with the corresponding months in 2004. In an accounting sense one could attribute all of this to the hurricanes, which resulted in an average of 1.1 mbd of shut-in production from the Gulf of Mexico in September and October. The more complete part of the story, however, is that production increases in the rest of the world failed to materialize to compensate for the lost U.S. production. This fall we were thus seeing the confluence of two forces: demand was declining as consumers adjusted to the higher prices, and supply was declining as a direct result of the hurricanes.
But the need to keep global demand in check remains an ongoing reality. There remains the basic difficulty of reconciling surging demand from China and other developing countries with the practical geological challenges of continuing to increase global oil production. The bumps from each day's news are important, and could ultimately prove to be events of quite profound impact. But those operate within, and in some ways are themselves manifestations of, a broader context that should not be overlooked. As I stated earlier, I don't expect a collapse in oil prices even if peace settles like a dove on Iraq, Iran, and Nigeria.
And I'm not exactly counting on that to happen.
Posted by James Hamilton at January 18, 2006 08:45 PMdigg this | reddit
Listed below are links to weblogs that reference Oil market jitters:
» Checking in on oil prices from Houston's Clear Thinkers
While the price for natural gas has receded quickly from the post-hurricane highs of last summer, oil prices have been a different story. Crude-oil future contracts on the New York Mercantile Exchange climbed about a dollar yesterday, settling at a... [Read More]
Tracked on January 20, 2006 04:41 AM
» This Week's Carnival of The Capitalists from Financial Rounds
This week's COTC is up at Patent Baristas. As usual, there's quite a diverse assortment of posts. Since my focus is finance and economics, there were a few that particularly caught my eye: [Read More]
Tracked on January 23, 2006 09:15 AM
I notice that other mined commodities as well as oil are also at long term highs - copper, zinc etc. No-one (as far as I am aware) is making the case that these commodities are in short supply in the ground or have become more expensive to find, instead it is a lagged response to a sudden acceleration in demand (from BRICS). That is what happens in highly capitally intensive industries with long lead times in adding capacity. I believe the same is happening with oil, I would be more convinced, in other words, of the peak oil scenario if oil was the only commodity at high price levels. The short term ups and downs in supply that you describe are to be expected as new capacity is a large part of the total supply since, as any engineer knows, you get the bugs out a system in the early years of operations. Another point is that the oil industry has changed - in the not too distant past capacity could be easily added in small chunks by many small individual operators, now it is added in very large chunks by a few large operators. This is making the supply curve a lot more "stepped" versus the smooth curve of the past, plus more "lagged" in time - both contributing to volitility.
Posted by: ChrisA at January 19, 2006 03:50 AM
The "peak oil scenario" suffers IMO by being narrowly defined, differently, by different people.
I think the broad move toward tar sands and oil shale says something about the availability of convenient liquids. Similarly the effort to build natural gas terminals says something about the availability of convenient gases.
That makes "peak oil" in the broad general sense true, for me, but I realize that for other people "peak oil" means we are all going to freeze tomorrow, or whatever.
Posted by: odograph at January 19, 2006 04:57 AM
ChrisA -- over the last 20 years the price of oil has a 0.97 correlation with the price of copper and if you lag it one month the price of oil this month has a 0.93 correlation with the copper price in the prior month.
Over the longer run the correlation is not that strong, but copper still is a very good leading
indicator of oil prices as is the CRB raw materials index. Both tend to reflect world economic growth and demand.
Your other comments are very good.
Posted by: spencer at January 19, 2006 08:57 AM
Isn't it possible that the change in price of the other commodities is reflecting the deflation of the dollar caused by the oil supply/demand changes? Heck, some people have fairly good economic models where the currency is BTUs or barrels of oil or whatever.
Posted by: M1EK at January 19, 2006 11:14 AM
Wouldn't your want to regress changes in oil prices on lagged changes in copper prices? My guess is that autocorrelation dominates the levels regression.
Posted by: toddleem at January 19, 2006 12:18 PM
More on oil price jitters.
I often look at the price/cash flow, price to nav. etc. of North American Gulf of Mexico based exploration and production companies (cheaper drilling costs in shallow water, but no big finds, lower reserve life, i.e 7 years) versus onshore companies (more costs, longer lived reserves-i.e. 14 years)
A basket of GOM stocks: EPL, PQ, REM, SGY, ATGP, BDE, and WTI sells for 3.0x 2006 estimated cash flow. The onshore group-CRK, RRC, SM, BEXP, PVA, and SFY sell for 5.3x.
The historical ratio for these two groups is about 75%. We are at 56% now.
While this ratio might express hurricane fears, I also think that it is a proxy for reinvestment risk-the market doesn't want to put a multiple on low reserve life. It seems to me that this ratio is a "fear indicator" and something else we can look at when measuring supply fears.
Note that most of the above mentioned stocks are small cap stocks so reserve additions ought not to be that all that tough. The small GOM guys ought to be able to develop small fields, take up stranded fields, etc. They don't need home runs to grow reserves, simply a lot of singles. With larger companies, this ratio may not be as relevant since it may truly represent the difficulty finding home run reserves.
Posted by: toddleem at January 19, 2006 12:58 PM
I agree that the recent prices of oil are determined mainly by increasing demand. But price is always the result of both supply and demand. In the '50s and '60s oil consumption increased at a very high pace for many years - with very stable or decreasing oil prices. What has changed is the supply situation. Oil depletion is a real factor.
I don't subscribe those scenarios where peaking oil production means extremely high oil prices. It is not only supply or only demand. The mechanism is mostly indirect and more subtle. Look what we are seeing now in Europe. There have been some natural gas disruptions and electricity supply problems recently for various reasons, now because of very cold weather. The prices do spike, but the spikes are promply cut by industry decreasing its consumption. So prices don't explode and households don't feel the crunch. Only industrial production is down.
Repeating supply problems will lead to diminishing industrial investments and outsourcing of production. And this is happening right now. It is not very dramatic but a very large scale move. China is not only cheap labor, it is also the only industrial country where domestic energy production can grow about 10% every year. Nobody can match this.
Oil prices can fall considerably if we get a very serious recession. And we will get a recession if the world energy production growth will stall.
Look, I know that you don't subscribe my "Chinese coal" theory. But may be it is the different viewpoint. We in Europe have recently seen some experimental macroeconomics in huge scale (and we can count Soviet socialism here, too). In Eastern Europe all market forces were let completely loose without no regulation. It didn't work, but population diminished. In the EU there was a try to create larger markets and more competition in a very dense regulatory framework. The economic growth has stalled. No success here, either.
Experimental macroeconomics is fine and we should learn something of this. The results were obtained at huge human cost. The main point is that those experiments were based on the mainstream economic theory. Well, we have tested the theory. I think the problem is that the theory is not wrong but inadequate. Newtonian mechanics work but only in certain context. Neoclassical economic works but only in certain conditions.
Posted by: TI at January 20, 2006 12:19 AM
this just in
not directly related to the discussion of oil........
or is it?
Posted by: kuros at January 20, 2006 05:53 AM
Yes, of course it is. I was talking about the demand factor here. Oil and other commodities demand and thus pricing is dependent critically from the Chinese economic growth and the Chinese growth depends mainly on the growth of coal production. This is the connection. This is to say that oil prices are less dependent on - say - from FED rates.
But we could look closer to the oil future prices formation and ask, for instance, what is the impact of speculation here. I would guess the "speculation add-on" has this year been on average about $10/barrel. This has made oil price fluctuate between $60-$70, the basic line rising.
Posted by: TI at January 20, 2006 09:51 AM
Thanks Spencer for your comments. I am surprised the correlation of oil to copper is so good, I would have thought that individual industry cycles would not be that related, except in the case of a demand shock.
Toddleem - could the difference in stocks PE be related to the proportion of gas to oil. I would guess that gassy stocks are more valued than oily stocks by the market because the market sees more of a moat around gas prices than oil. Just a guess.
I don't really want to get into a debate about whether we are at peak oil except deal with one point made by TI about the rapid demand increase in the 50s and 60's not resulting in high prices in oil (and other commodities. What is happening now is not just increasing demand but an acceleration of increasing demand. Industries can plan and "tool up" for steady increases in demand, acceleration in demand is much more difficult to plan for. Stephen Roach talks about the growth wedge from BRICS and how their share of annual global growth starts to become very significant post about 2010. Effectively this will be seen as an acceleration in world economic growth. Think about the world as two economies, the first area is large but growing slowly say 1%, the second area is small but growing fast , say 10% - initially total growth of both areas together will be 1%, eventually it will be 10%, in between there will be a period of rapid acceleration from 1% to 10% as the fast growing area becomes a similiar size to the slow growing one. I think in commodities (because they are used in more basic industries), we are already past that milestone. Of course nothing ever grows to the sky, so the party will be over sooner for commodities than for global growth.
Posted by: ChrisA at January 20, 2006 11:34 AM
If we look at the oil statistics we'll see that oil consumption grew much more rapidly in the '50 and '60s than ever later. During the '00s oil consumption growth has been less than 3% on the average. Before the first oil crisis 1966-1972 it was over 7% in the average. And the growth rate did accelerate during the '50s and '60s somewhat. If there is an acceleration tendency going on at present, it shows only in the couple last years, so it is too early to say anything about it yet. So the reaction of oil prices to the demand growth is different this time.
Otherwise Chris is right to point out the effect of BRICS in the global growth. My point is that behind this growth is increasing domestic energy production in those countries. Especially China and also India have shown high domestic coal production growth - but China far higher than India. Domestic coal production correlates quite well to the growth rate.
Posted by: TI at January 21, 2006 01:02 PM
On copper-oil price correlation: copper is one of the most energy intensive commodities, so there is at least some real connection to the energy prices. But this doesn't explain wholly the close correlation. More important factor may be copper depletion. World copper production doesn't cover demand. Copper production grew about 4% during 2005, or about at same pace than oil production. Chile, worlds most important copper producer announced in November that its production will be lower in 2005 than 2004. Has Chile copper peaked? World copper production growth stalled about 2000, but the demand has risen - China figures also here.
The very close correlation must be mostly a coincidence. But oil and coal are both minerals and are subject to same kind of depletion patterns - and just happen to be in same kind of situation. Besides they must have quite similar demand elasticities and therefore react same way to economic growth. Anyway, the oil-copper correlation is an interesting one.
Posted by: TI at January 21, 2006 01:49 PM
About "peak copper": there is an interesting article in Scientific American.
Posted by: cerqueira at January 22, 2006 06:12 PM
I tend to agree with Julian Simon I am afraid in terms of doomesday scenarios for commodities - see chapter 2 - it actually discusses copper as an example.
Remember this was written over 20 years ago in response to similiar fears.
On the statement that the current high prices imply a peak for both oil and copper, by that logic you must also say that iron ore, zinc, silver, gold etc are also at a peak (since they are all at high price levels as well) - this would be more than an incredible coincidence - all these resources depleted at the same time? I submit that the more likely cause is a demand shock.
On the question by TI of the high increases in rates of consumption of oil in the 50s and 60s not leading to high prices as today - this is again getting to the heart of what is meant by a demand shock. An acceleration in demand rates from 5% to 7% per year is actually, counter-intuitively, going to cause less of a price impact than an increase from 1% to 2% a year. The reason is that any commodity based industry actually consists of two industries. There is the operating part (which operates existing capacity and sells the product) and the new business development part (the piece that creates new capacity - in the case of oil this is everything from the Oil Company's exploration department, drilling rigs, fabrication contractors etc). In the case of an demand increase of from 5% to 7% the new business development part has to increase in size by 40% to keep up with demand, in the case of a rise from 1% to 2%, the new business part has to increase by 100% in size. Tremendous price signals are required to double to size of an industry, especially when the commodity industry has got used to low margins and low growth rates as has happened in recent years.
Posted by: ChrisA at January 23, 2006 05:04 AM
Note that one of the above stocks, Remington, was taken over today at a 20% premium. Other gulf plays are running up in tandem. Perhaps what I am looking at here is simply a market inefficiency-the market likes the sexy story in place of the dull cheap story.
Posted by: toddleem at January 23, 2006 06:45 AM
excuse me, but upon reading some parts of the Julian Simon website that you referenced, i can only conclude that he is a complete idiot.
Posted by: sampo at January 23, 2006 11:03 AM
I always find the discussions about the bet between Simon and Ehrlich to be interesting. Ehrlich lost that one, and that result is what we remember. However, if we go back and look at prices of the commodities chosen, it is apparent that the timing of the bet is crucial. Had they made the bet for the time period 1970-1980, Ehrlich would have won. Ditto for 1985 - 2005, for example.
Posted by: brecha at January 23, 2006 06:31 PM
Chris, oil exploration doesn't work that way. Oil production is extraction and and oil fields in production are constantly depleting. That is why there is always a lot more exploration and development going on than just actually increasing the production. An increase from 1% to 2% doesn't involve doubling the new exploration capacity - in fact it might not involve any increase in that capacity because that little increase might come from supply reserves already at hand.
Besides the '50s and '70s production rates didn't increase smoothly but were quite high all that time. No, it is really geology here.
And the copper: all commodity price have risen, but copper has the closest correlation to oil. This is the difference. CRB index has risen about 25% from 2004 to present, but oil has nearly doubled, and so has copper. So we could say that oil prices have the "commodity demand" component, but that is only about 25%. The rest may be "tight supply" component.
Posted by: TI at January 23, 2006 11:18 PM
Oh dear, I said did not want to join a debate about peak oil, and here I am doing just that! I will therefore make this my last contribution.
As you know the CRB index contains many other non-mined commodities eg orange juice, cotton, cattle etc so we should not expect it to be as affected by the demand shock as I describe it. However I do think it weakens your case since it is also at 30 year highs - does this mean that we have reached peak orange juice?
The arithmetic I provided to show the impact of demand shock on the supply chain was illustrative, not intended to be specific to one industry. But it still works even if you include a component to account for base decline. The numbers are simply less dramatic. My point is that an x% increase in demand has to be compared against the new business development capacity of an industry not the installed productive capacity. Using the oil industry as an example - the drilling industry (which is part of the new business development capacity) is currently running flat out - drilling rig costs are at well beyond new build levels - if I could wave a magic wand and create a host of new rigs today they would be utilised. The cause of the current high oil prices is because the oil industry has only invested in enough to add a certain amount of new capacity per year (because that was all that was needed until now), and the current increase in demand is above that level. Eventually the industry will add more capacity, just in time probably for the BRICS to move to a more developed country growth path so collapsing prices. This is a demand shock not a supply shock.
If you are still reluctant to believe in the devastating impacts of perturbations on supply chains I recommend that you play MIT's Beer Game - you will be amazed how wrong even a small simple supply chain can go.
On the Simon and Ehrlich bet I think the timing was all to the point - at that time there was a similar scare as today because commodity prices were high - Ehrlich's view was this was due to depletion - the real problem at that time was a supply shock and Simon's view was that human ingenuity would take care of that problem. He was proved right.
Posted by: ChrisA at January 24, 2006 04:29 AM
It is probably possible to model whether such reinvestment (in oil and gas particularly) now yields the same production increase it did in the past.
I read in Peak Oil sites, for instance, that we are drilling a lot more in North America for natural gas, but that production continues downward.
I wonder, if "ingenuity" was really the savior in past cycles, rather than the ability to move offshore? I guess you could count something like North Sea oil either way (as innovation and a new reserve).
Unfortunately, while the North Sea technolgy carries on, the North Sea reserve does not.
Posted by: odograph at January 24, 2006 06:40 AM
I’ve read many articles on the internet and in trade publications indicating the price cycle of crude oil may extend over several years, responding to consumer demand along with OPEC and non OPEC supply issues. This is important information that residential oil consumers must take into account when choosing a fuel oil dealer.
Omni Marketing Services is a contract telemarketing company that works with residential oil companies on a business to consumer level. We have noticed a trend within the last six months of an increased number of C.O.D. oil consumers migrating to Full Service, looking for price protection in the form of a capped or fixed price on their oil.
As we approach winter we can expect oil prices to spike as oil supply falls short. Consumers that have locked into a price protection plan will save a substantial amount of money over the course of the heating season.
There are many advantages in doing business with a Full Service Oil Dealer :
1. With a Full Service Dealer you are guaranteed oil, no matter how bad the winter gets. With a Discount Dealer, if the dealer runs out of oil… you’re out of luck !
2. With a Full Service Dealer you receive automatic oil deliveries. With a Discount Dealer you need to be diligent and watch your oil tank gauge, calling when you think you might need oil . Many consumers have run out of oil because of this.
3. With a Full Service Dealer you will be on a convenient budget plan that will allow you to pay by check or credit on a monthly basis. With a Discount Dealer you either wait for the oil delivery and pay the driver or you tape your cash or check to the fill pipe or front door.
4. With a Full Service Dealer you have a team of trained burner technicians that are on call 24 hours per day, 365 days per year. With a Discount Dealer you will have subcontracted technicians that may or may not respond to your call.
Given the extreme volatility of oil prices, along with the advantages listed, it only makes sense that consumers are choosing Full Service Oil Dealers.
Posted by: Omni at January 24, 2006 02:12 PM