Monday, September 29, 2008

Oil, Speculators, and Inventories

Whenever the price of something changes in a way some people don't like—up or down—someone blames it on speculators. Recent rises in the price of oil are no exception.

In some cases it is true; speculators can affect prices. They raise prices by buying goods and storing them, thus reducing the total amount available to be sold. They lower them by selling goods from their inventory.

As best I can tell by a little casual googling, world storage capacity for petroleum comes to less than two months output. At any given time, much of that is already in use. If, say, 80% of storage capacity is already being used, which I gather is a not particularly high figure, then the most speculators can do is to take off the market something less than two weeks worth of output—once. It is hard to see how that could have much effect on price for more than a short time.

There is, however, a cheaper way of storing oil. The reason to buy and hold oil is the expectation that you will be able to sell it in the future at a higher price. That is also a reason why a producer with limited supplies might choose to pump next year instead of this year. Storage capacity in that form is essentially unlimited; producers could choose to pump no oil at all and leave all of it in the ground. That suggests that, insofar as speculation is responsible for current high prices, it is speculation not by speculators buying oil and putting it in tanks but by oil producers leaving it in the ground today so as to pump and sell it tomorrow, or next year, or next decade. I do not know enough about the evidence on actual and potential output to guess how likely that is.

One of the odd features of the current political fuss over all of this is the widespread assumption that high prices due to speculation are a bad thing. That assumption seems to be shared by, among others, most of the people who argue that we are running out of oil and so should use less of it. If we are running out of oil that is a good reason to shift consumption from the present to the future, when it will be scarcer. Which is exactly what speculation, by producers or speculators, does.

12 comments:

Anonymous said...

The "speculators" are buying oil futures. But the contracts are settled in cash; they never take delivery of oil, so it's not clear to me how (or if) they can affect the spot price.

David Friedman said...

In response to Anonymous:

I think the only way they can affect the spot price is if they bid up the future price and that causes some producers to postpone production in order to sell their oil in the future at a higher price. Or people to store oil, but I'm arguing that's a small effect.

Anonymous said...

Delaying production of oil (or gas) is technically feasible, but in many cases it's not practical.

Taking one or many wells offline can be costly, it is not like merely flipping a switch, and various state and federal regulations commit the producer to certain production quotas as a condition of being granted a production lease. At least in this country, I don't know if other countries operate similarly.

Given my familiarity with the industry, I would expect that speculation on oil (& gas) price is more likely to effect a company's near term exploration strategy, and possibly their extension development plans, but not, I think, their current production quotas. Although choosing to spend less on exploration could achieve the same effect of leaving the oil in the ground.

Anonymous said...

David,

I recently finished reading The Machinery of Freedom, and I found it absolutely fascinating. You paint a very compelling picture of how an anarcho-capitalist society might work, and have convinced me that it is not only desirable, but practical too. Thank you for writing it, and I'm very much looking forward to reading Future Imperfect, which I'll be ordering soon.

Anonymous said...

Actually, I agree with almost all of this. "Speculators" in oil can take advantage of, and cause, price fluctuations on the order of days or weeks, but not years. The multi-year trend is simply Up.

God isn't making any more oil. There's a finite amount of it in the Earth, so in a free market, the price would rise as supply vanishes, forcing demand down to match supply; this would continue steadily until supply and demand both approach zero and price approaches infinity. (Yes, price has to keep going up, or demand won't go down.) A rational producer strategy, then, is to hang on to your reserves until later, when they're more valuable. (Some oil companies may indeed be doing that, although the U.S. government is trying to do the exact opposite.)

For that matter, a rational consumer strategy is to reduce consumption every year by at least the percentage of remaining reserves that was used up in the past year. But that's a tricky calculation to make, and if you don't reduce by quite enough, prices shoot up, there are shortages, etc. So if a government were to intervene at all, it should be to artificially raise the price and lower demand, so the annual reductions are an overestimate rather than an underestimate, and we guarantee a soft landing.

As an environmentalist, I'm all in favor of high energy prices: they encourage energy-efficient industries and habits, and discourage inefficient ones. The sooner that happens, the softer the landing will be.

Raphfrk said...

In theory, the price should rise at the same rate as the market. If the market is rising slower, then it is worth reducing pumping as you get more money for the oil next year than selling it now and investing the money.

This reduces supply this year and raises it next year, increasing prices this year and reducing them next year, so the price rise is moved closer to the increase in the market.

Ofc, that assumes endless increase. If technology will give an alternative to oil at a certain price, then this is the max possible price of oil.

If this new development is going to happen soon, then it is worth pumping at max now, as the high price of oil won't last forever.

David Friedman said...

"In theory, the price should rise at the same rate as the market."

I assume that's supposed to be "as the market interest rate." That result depends on simplifying assumptions--in particular, ignoring the cost of extracting the oil. What should go up at the interest rate is the net revenue to the producer per barrel of oil pumped, which is price minus cost of pumping it. A second assumption needed is perfect foresight--owners of oil know what next year's price is going to be.

That result, incidentally, comes from Harold Hotelling's analysis of the theory of depletable resources, done about seventy years ago; unfortunately most people other than economists are unfamiliar with it.

Don Geddis said...

"[A] producer with limited supplies might choose to pump next year instead of this year. [...] I do not know enough about the evidence on actual and potential output to guess how likely that is."

The Peak Oil folks do. (E.g. theoildrum.com.) Month global production, on a country-by-country (and region-by-region) basis, is tracked extremely closely.

So far this year, global oil production is at an all-time high. There has been a lot of discussion for a few years that global oil production might have already peaked in 2005. (In May 2005, to be precise.) Global production in 2006 and 2007 was less than 2005, and every month was less than the month of May 2005.

But then Jan 2008 somehow (!) broke the May 2005 record. 2008 is shaping up to be a global high in oil production, although we'll have to see what happens in the last half of the year with the US financial crash.

But in any case, you've correctly shot down the argument that oil speculators are driving the wild spot oil price swings, using only trading on oil futures (since they don't seem to be storing it).

Similarly, Peak Oil data will show you that the producers aren't keeping it in the ground either. Sure, I suppose hypothetically someone might have been able to produce more than they actually did. But it's highly significant that the world did produce more oil in the first six months of this year, than in any six month period in history.

So the theory that price shocks are due to producers deliberately holding back production, is also a broken theory.

Unknown said...

hudebnik, you are correct that there is a finite amount of natural resources (carbon-based fuels take millions of years to form and certain biosystem). However, what's relevant is that 1) proven reserves are being continuously discovered, 2) as prices of the resource rise the feasible supply pool increases (there are different chemical processes with different costs that get the same output -- e.g. tar sand boom in Alberta, drilling the continental shelf in various previously-uneconomical places), 3) new technologies allow us to make more output with a fixed amount of inputs (modern computers vs. cold war-era computers are an extreme example but recycling also comes to mind). These factors tend to offset the "finiteness" factor.

Also relevant:
http://en.wikipedia.org/wiki/Simon-Ehrlich_wager
http://www.env-econ.net/2005/09/the_simonehrlic.html

Finally, reducing current consumption by ad hoc % quota is not optimal if infinite discounted stream (future use is slightly discounted + strictly increasing utility) of consumption is to be maximized. When the price of oil rises sufficiently high that substitution to other resources becomes optimal - the age of oil will be over well before "we run out."

Anonymous said...

What is your estimate that speculation has had on oil prices?

Andrew said...

Hudebnik wrote:There's a finite amount of it in the Earth, so in a free market, the price would rise as supply vanishes, forcing demand down to match supply; this would continue steadily until supply and demand both approach zero and price approaches infinity.

You're making a huge assumption, that is that there is no substitute for oil. Gasoline in california already includes 5-10% ethanol. We can create oil by coal gasification at the equivalent of around $5 a gallon. We can create oil from algae for around $40 a gallon equivalent. As technology improves these costs may go down.

Transportation (accounting for 70% of oil consumption) in the next ten years will go through an electric revolution. With current battery technology, electric vehicles are already approaching the total cost of operation of equivalent gas vehicles (see Tesla, Chevy Volt, plug-in Prius, Chrysler EV, Chinese and Indian EVs), and in ten years i predict that the majority of new vehicles will be primarily fueled by electricity. That will create a huge drop in gas demand.

I would not be surprised if the $140/barrel oil we saw recently would be the highest inflation-adjusted prices that we'll see in our lifetime for oil.

Anonymous said...

Vadim Iaralov writes:
1) proven reserves are being continuously discovered,

Which doesn't change the total amount of oil in the slightest; it only changes our estimate of the total amount of oil. The discovery of a new oilfield can have only a short-term effect on the overall price/supply/demand scenario I described.

2) as prices of the resource rise the feasible supply pool increases (... e.g. tar sand... drilling the continental shelf...), 3) new technologies allow us to make more output with a fixed amount of inputs ...
These factors tend to offset the "finiteness" factor....


And Andrew writes:

You're making a huge assumption, that is that there is no substitute for oil....


I am certainly not assuming that. When I say "demand", read it as demand for oil, not for its substitutes; substitution is presumably one of the ways demand would drop, along with efficiency ("make more output with a fixed amount of inputs").