January 17th, 2018

ICONS Home :: Archives :: Contact  

Useful Knowledge For Your Next Conversation about Oil

Professor Ferdinand E. Banks
February 19, 2008
The University of Uppsala, Uppsala Sweden

What I want to do in this short note is to systematize, on a non-technical level, the most important aspect of short-run pricing in the oil market. You already know something about this topic if you read the (UK) Financial Times, or watch CNN and/or Bloomberg.

A few years ago, in the Financial Times (November 4, 2004), the commodities page contained the following information. "Crude oil futures moved lower in volatile trade, following a large increase in U.S. commercial crude inventories, signalling that the oil market is well supplied." Both CNN and Bloomberg - especially the latter - constantly mention inventories, but this kind of terminology is completely absent from mainstream microeconomics textbooks, and the same is true of most books on energy economics. This is one of the reasons why many academic discussions of the oil and gas markets are largely without any scientific value.

The first thing to take notice of though is a basic long-run consideration. When there appears to be a durable increase in the oil price, it usually means raising the reference price that is used for capital expenditure, which in turn is an important determinant of long run oil supply. For example, a few years ago an oil price in excess of $40/b resulted in BP raising its reference price to $20/b from $16/b. Thus, in one of those famous discounted cash-flow exercises, $20/b would be inserted as the ex-ante expected future price of oil over some time horizon. Marathon Oil and ConocoPhillips were apparently using $21/b, while Statoil had raised its figure for a sustainable oil from $18/b to $22/b. In examining these figures, readers should get an idea of the changes that have taken place in this crucial market, since ex-post the situation is totally different.

A familiar claim at that time was that these were substantial increases, although lower than those I suggested employing. There is no point in making heavy weather of all this, because at the present time 'Big Oil' is showing great caution about the 'rate of return' liable to be realized on available projects, despite the substantial increase in earnings because of the high oil prices. There is, for example, a constant mention of the lack of high quality drilling prospects. This has led to an increase in share buy-back arrangements, as well as higher dividends for shareholders, instead of the expected large increases in exploration and production budgets. Where investment spending is increasing, it often tends to be concentrated on obtaining more oil from older fields.

This makes economic sense, given that everywhere in the world oil industry exploration programs are turning up fewer reserves than they did on average during the l990s. There are also some unexpected disappointments, with the latest apparently being those experienced by the big Russian-BP project on Sakhalin.

We can now look at some elementary analytics of short run pricing. I will make a few introductory remarks which focus on an original diagram. This diagram, which at first glance may appear difficult, is actually extremely simple. It should be examined by all readers, and become a part of their conversational equipment! I can also mention that students in my course on oil and gas economics in Bangkok who could not reproduce and discuss this diagram in the final examination were assured of a failing grade.

Oil inventories (i.e. stocks) are a stock concept : they are defined in e.g. barrels, and measured at a certain point in time, but they lack a time dimension. In the discussion below they are designated by AI and DI. On the other hand, current production (s) and demand (h) are flow concepts: they are defined and measured in terms of a certain unit of time (e.g. mb/d). Microconomics 101 deals almost exclusively with flow variables. Stocks and flows are closely related, since the change in stocks is determine by the net investment in stocks during a given period, or s - h: i.e. flow supply (s) minus flow demand (h). Moreover, we define equilibrium in the stock market as the situation where desired stocks (DI) are equal to actual stocks (DI = AI), where that famous concept equilibrium can be defined here as 'a state of rest' (or, perhaps better, outcome = intentions). If the stock market is out of equilibrium, e.g. DI > AI, then in the flow market we must have s > h in order to fill this gap. In order to obtain this price must increase (to raise output over consumption), and the amount of the increase says something about how rapidly inventory holders want additional inventories. We can therefore write the price change (per period) as Δp = v(DI - AI), where Δ is the difference operator. What we have here is a simple linear relationship between excess stock demand and the price change, with v is a constant. Now let us look at the diagram.

The current (or flow) supply (s) goes into stocks (i.e. inventories) and current (i.e. flow) demand (h). Price is formed by the relation of actual stocks (AI) to desired stocks (DI), with the flow equilibrium [s(p) = h(p)] playing a secondary (but important) role. The equilibrium expression is AI = DI, and when this situation prevails, s = h, and price is constant! Put another way, a stock equilibrium implies a flow equilibrium, while a flow equilibrium does not imply a stock equilibrium. In this type of model expectations are very important because of their influence on desired stocks, and in the real world expected prices are undoubtedly more difficult to describe than via the simple expression shown in the figure: pe = f(p). Let me also mention that in my energy economics textbooks I include a more conventional analysis, employing stock and flow supply and demand curves of the kind developed by Bushaw and Clower (1957). These two authors also present a comprehensive mathematical treatment of the kind of dynamic problems that are implicit in stock flow models.

Returning to our expression for a change in price, which was Δp = v(DI - AI), with 'v' designated a positive constant in this elementary analysis. It is thus obvious that when desired inventories are greater than actual inventories, price increases. This leads us to ask how we reach an equilibrium, and the answer is supremely uncomplicated. If the flow demand and supply curves are of the usual types - i.e. have the usual slopes - then the increase in price raises flow supply above flow demand, resulting in an increase in inventories that continues until AI = DI, and Δp =0.

Readers who want a formal outline of the mathematics associated with the above diagram can examine my forthcoming paper 'Economic Theory and the Price of Oil', but the thing to be aware of now is that the price could be extremely volatile. Students of electrical engineering should immediately note that the diagram (with its feed-back 'circuit') fits the description of a servomechanism, in which case very high volatility is perfectly natural. This volatility is one of the reasons why futures and options markets for oil are so important. Occasionally they cause financial headaches for careless transactors, but in general these two markets have functioned extremely well.

Perhaps a few more comments about futures markets are justified. Hopefully, readers are familiar with the elements of hedging and speculating, but if not let me refer again to my new energy economics textbook, which contains materials that I presented in my elementary courses in energy economics in both Stockholm and Grenoble.

In l980 the circumstances that make a futures market attractive to speculators appeared in full flower. By that I mean an increased volatility in the oil price. Furthermore, as the futures markets for oil products (and later oil) began to expand, they reinforced the inclination toward fewer long-term transactions in physical oil. Even with e.g. a high probability of falling oil prices, many buyers of oil felt that they should lock in present prices by a judicious use of futures contracts, and so they negotiated a smaller amount of their requirements through the medium of long-term contracts. (The key issue here of course is not just 'probability', but degree of risk aversion.)

Analogously, there was a decreased propensity by some categories of buyers to hold inventories as a precaution against having to buy in a rising market. Instead, individuals and firms requiring oil in the future bought a number of futures contracts. Later they went into the spot market to buy their oil, while at the same time making an offsetting sale of futures contracts. As noted above, the oil futures market has functioned in an ideal manner, and the same is true of the markets for oil options, oil futures-options, and oil swaps. Without going into the matter here, the presence of these derivative markets has contributed to restraining rather than increasing the price of oil, since they have reduced the risk of buyer and sellers.

The oil derivatives markets may also have upgraded market efficiency by providing an increase in the quality and quantity of information. This has probably helped markets with a large number of participants, having various degrees of risk aversion and many opinions about future trends or tendencies, to be quickly brought into equilibrium. Just as important, expectations concerning the future movement of important variables can be continuously appraised in a quasi-scientific manner. As a result, the prices displayed on futures markets might be instrumental in forming expectations about future prices, although it should never be forgotten that futures prices are hardly more than rough estimates of spot prices in the future. Someone who apparently has failed to fully comprehend this reality is the present Chairman of the U.S. Federal Reserve System. (And notice the emphasis placed of the words may, probably, quasi-scientific, might and rough! Readers are invited to brood over these.)

Now that we have looked at the benefits accruing from futures markets, something should be said about the costs. It can be easily shown with a few simple difference or differential equations that the information provided by a futures market might have a destabilizing effect on spot prices due to its causing rapid shifts in the demand for stocks. There is perfectly logical, since if inventory behaviour is influenced by changes in expectations about futures prices, any institution that encourages frequent revisions of expectations can be identified as a potential source of increased price volatility. Just what weight should be attached to this defect is uncertain, although it might be possible to argue that volatile prices could increase the attraction of futures markets for speculators, and so the increased amount of liquidity might make it possible to support more hedging (i.e. insurance against price risk). But this does not mean enough liquidity to support extremely large-scale hedging by the major oil firms.

The way I deal with the efficiency or otherwise of the oil futures markets is to look at the bottom line. As far as I can tell complaints against those institutions are few and far between, and I feel comfortable with the contention that the presence of oil futures and options markets is better for almost all of us than would be the case if they did not exist.


Banks, Ferdinand E. (2007). The Political Economy of World Energy: An Introductory Textbook. World Scientific: Singapore, London and New York.

______.(2004). 'Beautiful and not so beautiful minds: an introductory essay on economics and the supply of oil'. OPEC Review (March).

______ (2001). Global Finance and Financial Markets. World Scientific: Singapore, London, and New York.

______ (2000). Energy Economics: A Modern Introduction. Kluwer Academic Publishing: Boston, Dordrecht, and New York.

______ (1980). The Political Economy of Oil. Lexington Books: Lexington Massachusetts. Bushaw, Donald and Robert Clower (1957). Mathematical Economics. Irwin Publishing: New York.

Cairns, Robert D. and Graham A. Davis (2001). 'Adelman's rule and the petroleum firm'. Energy Journal (Vol. 22-3:31-54).

______. (1998). 'Simple analytics of valuing producing petroleum reserves'. Energy Journal (Vol. 19-4: 133-142). Hicks, John R. (1939). Value and Capital. Clarendon Press: Oxford. Salameh, Mamdouh G. (2004). Over a barrel. Salameh: Surrey England.

Samuelson, Paul Anthony (1958). Foundations of Economic Analysis. Harvard University Press: Cambridge Massachusetts.

Professor Ferdinand E. Banks
February 19, 2008
The University of Uppsala, Uppsala Sweden

Home :: Archives :: Contact  


January 17th, 2018

© 2018