The relation between oil prices and economic activity
From Lewis L Smith
Recently a longtime Internet correspondent of international repute requested my opinion on the relation between oil prices and economic activity. I was both honored [since he is better known than I am] and somewhat take aback, since the subject is neither an easy one nor one on which I have done any research.
Fortunately I have tried to keep up with the work done by others and also have written quite a bit on closely related matters such as peak oil, “wild cards in the oil deck” and such like. So I was able to piece together the answer which follows, with a few revisions here and there. ==============================================================
The relationship between oil and economics is a paradoxical one.
On one hand, the production functions used by most economic theories are net of purchased inputs, so energy in any of its many and varied forms is nowhere to be found. Moreover, until 1973, coal, natural gas and oil were so cheap and the long-run supplies believed to be so abundant that their future hardly seemed worth worrying about. In fact, both Saudi Arabia and the oil industry continued to propagate the myth of this country as a cornucopia of crude oil right through 2004.
[This “balloon” was “punctured” in 2008 by none other than the King of Saudi Arabia and top officials of Saudi-Aramco, the national oil enterprise. He said in effect that future discoveries will be for the Saudi children and not (by implication) for your SUV ! And the top officials said that Saudi production will peak within a few years.]
On the other hand, since 1973, the impact of fluctuations in oil prices on the rhythm of economic activity [and vice versa] has been impossible to ignore. So in the arena of applied economics, the interaction between the two has generated a large literature among economists and other concerned professionals, some of which is contradictory and/or acrimonious !
In addition, the debate is complicated by issues of methodology as well as of substance.
For example, a seldom-mentioned issue is the use of consumer price indexes to deflate economic data recorded in current prices, The fact of the matter is that many of consumers of petrofuels are industrial enterprises at some distance from the final consumer, so the use of another type of price index would seem to be indicated.
For example, a great many people insist on dividing reserves [what remains to be extracted from a reservoir of oil or gas] by current production, to obtain an estimate of a reservoir’s expected life. In fact, the life-time production curve of any reservoir is “a hill” of sorts, so it only hits its average twice, once when production is increasing and once when it is decreasing. And active reservoirs which have passed their peak will last a lot longer than people think, but at rates of production of which decline fairly steadily. A good example is the Mexican oil field called Cantrell, which within six years has gone from roughly two million barrels a day to less than six hundred thousand. This is important because the timing of peak oil depends on optimum production much more than on reserves.
My impression of the present status of the debate over oil prices and economic activity is as follows —
 Frequently there is a relationship, albeit a complicated one.
 Causality can run in either direction. That is, an oil-price spike can help to bring on a recession, but a prolonged period of economic growth can provoke an increase in oil prices, as happened with the run-up which ended with WTI at $147 per barrel in 2008.
In the latter case, the principal “drivers” seem to have been the following —
[a] A history of prolonged economic growth in many countries.
[b] Anticipations of continued strong growth by both the Chinese and the Indian economies, regardless of what might happen elsewhere.
[c] Failure to anticipate the financial crisis which brought on the Great Recession.
[Aside — Since my second son and one of my nephews are the fifth generation of my mother’s family who work in the construction and housing industries, I believe that this crisis was foreseeable and preventable (without consulting an economist) . But that is a story for another day !]
 When changes in oil prices do impact the economy more than vice versa, the impact is frequently asymmetrical and nonlinear. That is, a dollar-per-barrel increase at one time will not have same impact [in an opposite direction] as a dollar-per-barrel decrease at another. Metaphorically speaking, what goes up does not always come down the same way.
 In any one situation, there is no one explanation of what happened. Typically there are several, with one predominating, and feedbacks complicating our interpretation of events at every turn. Supply and demand are almost always present, but they seldom tell the whole story.
Having hopefully answered the question, I would like to add that it is often much harder to forecast demand, prices and supply for the oil industry than it is for most other industries.
This situation is partly due to  through  above. However, there are other factors. The “bottom line” seems to be that the oil industry is rather different from most other industries and different in important ways —.
 Although industry statistics are not as bad as those for the wine industry, they are plagued with errors, omissions and bald-faced lies. For example —
[a] As is well known, estimates of a reserves of a country [or its reservoirs] should change at least annually, up or down, but change. Yet many oil-producing countries show the same estimates for reserves, year after year. Every one of those identical figures are obviously false.
[b] Some important producers are believed to have exaggerated their estimates of reserves by a substantial amount, in one case by 111% !
[c] Some production statistics remain constant [either in total barrels or barrels per day] for months on end. Obviously one or more of those figures are false.
[d] Even if all oil statistics were accurate [which they are not] , they would still substantially exaggerate the amount of both production and reserves available for export. The latter numbers are usually ignored but are more important than gross production or gross reserves. For example —
The pressure in most major reservoirs throughout the world is declining and must be maintained by the injection of CO2, natural gas, water or whatever. This injection and the subsequent cleanup of the crude brought to the surface takes energy, in many cases the equivalent of 10 to 15% of the crude oil produced.
The processes used to extract oil from shale or tar sands for example are even more energy intensive. In a worst-case tar sands operations, for example, process energy may require the equivalent of 35 to 45% of the oil extracted.
In many major producing countries, increasing amounts of crude oil must be diverted to supply local consumption and, in some cases, electricity exports to neighbors who have gotten behind in their system expansions. Some producing countries are even so foolish as to subsidize the price of gasoline !
[e] No data by reservoir has been published for OPEC countries since 1982. This information is regarded a matter of “national security”, and people have been killed outright while trying to obtain it. What the world is fed are estimates for an entire country, often prepared to justify production quota’s, conceal violation of these quotas or for other self-serving purposes. Entities like BP, EIA and the OPEC Secretariat dutifully compile the numbers without challenging their sources. The latter has even accused its members of lying !
 Expansion of capacity is more difficult than expansion in the typical manufacturing or service industry.
[a] When one needs more capacity for a certain model or make of automobile, one may pay overtime, add a shift or add another production line, in lieu of building a new factory.
But beyond a month or so, one cannot substantially increase the production from a reservoir whose production has been optimized, without damaging its long-term prospects. So when one needs more oil, one must apply enhanced-oil-recovery techniques to an active reservoir or go out and find a new one. The first is relatively cheap but of limited benefit. The latter can be hugely rewarding, but is chancy and very expensive .
[b] Oil is always found underground or undersea, often at great depths, and seldom reveals its presence by visible signs, such as surface seepage or the infiltration of water well. The possible presence of oil-bearing rocks must almost always be determined seismological techniques and then proven and delineated by repeated drilling. Even with modern technology, the odds of finding commercial quantities of oil are against the seeker, until the reservoir has been proven.
[c] The oil industry is a technologically complex industry, so advances do not come easily and often take years to bring to commercial fruition.
Indeed at the present time, we see no research in process which might have an impact on crude-oil capacity in the next decade as great as did the switch from 2D to 3D interpretation of seismographic data did in the last decade. Ditto for anything with a potential impact as great as the family of technologies which enabled Chevron bring in the “Jack 2” well, producing 6,000 barrels per day at a depth of 32,000 feet below the surface of the Gulf of Mexico, and do so without an oil spill.
[d] Oil prices fluctuate more than most industrial prices, and reservoirs often take three to ten years to reach full production after initial geological work has been completed. As a result, drillers for oil are easily scared off by dips in prices or economic activity. This is true even for large producing companies with deep pockets.
 All industries must worry about cycles, product competition, non-product competition, technological ambushes by competitors and many kinds of trends. However, to an unusual degree, the oil industry must also worry about a kind of ambush which we call the “wild cards in the oil deck”.
These include “game busting” or “game changing” events which might or might not occur, most likely without warning, and festering problems which everyone is either ignoring or which appear to have no solution. Examples are —
[a] An Israel attack suspected sites of nuclear activities in Iran.
[b] A Saudi Arabia grant to the Israeli air force of safe passage over Saudi territory for such an attack, as postulated by The New York Times.
[c] A flare-up of the Israeli/Palestinian conflict.
[d] A slowdown of economic growth in China severe enough to redirect widespread popular anger over local conditions against the central government and provoke a breakup of the country.
[e] The shortage of water in the Middle East reaches crisis proportions. [This is the region’s greatest problem but is largely ignored at present.]
[f] A worldwide depression following peak oil or gas.
 Oil markets are heterogeneous.
Some do look like and perform like the decentralized, competitive markets postulated as universal by mainstream [Neoclassical, Neoliberal] economics. Examples are the New York Harbor markets for No. 2 fuel oil, premium gasoline and regular gasoline.
But other, important markets are “non conforming”, like the markets for the major families of crude oils. The latter are characterized by —
[a] A concentration of production and reserves.
The “Oil Belt”, which stretches from Algeria through the Middle East, currently accounts for roughly 35% of worldwide production of crude oil plus refinery-type liquids derived from natural gas. Very roughly speaking, the Belt possesses 63% of the world’s crude-oil reserves. When some of its key members agree, it is much more effective than the Organization of Petroleum Exporting Countries. The latter is usually a “paper tiger” which cannot even return crude prices to the levels required by the public-sector budgets of its members, whenever these prices fall too low.
[b] Excessive gross margins
Since the beginning of World War II, the prices of the principle crude oils have defied a basic rule for competitive-market equilibriums. They have far exceeded the incremental cost of producing one more barrel of oil.
In the days of $2 crude oil, Saudi Arabia was clearly the supplier of last resort, but its incremental cost was something like 13 cents per barrel ! Today the situation is murkier. No one really knows who is the supplier of last resort nor its incremental cost.
Fortunately we do have some rough “ballpark” estimates for possible candidates. These are $70 per barrel for deep-water crude from Angola, $50 for unfinished oil from Canadian tar sands, $26 for US crude from new reservoirs, $15 for crude from the worst-case countries in the Oil Belt and $2 for Saudi Arabia.
Most of these estimates are obviously way below the market prices of recent years. As a result, the incremental gross margins realized by producing countries are not only positive, but the average gross margins, based on the difference between average market prices and average out-of-pocket costs, are very large for most of them.
[c] No equilibrium prices
Contrary to theory, there have been no equilibrium prices for any of the major “baskets” of crude oil during recent decades, only “equilibrium ranges”, ranges more often honored in the breach than in the observance.
In the case of West Texas Intermediate, a Western Hemisphere favorite, its price has spent more time outside its range than within it, and more time above its range than below it. In crude-oil markets generally, equilibrium prices are no more than occasional “pit stops in the road race of life”. And when they do appear, they are often “fragile and elusive”.
[For WTI, see Hammoudeh, Shawkat M., Ewing, Bradley T. and Thompson, Mark A. (2008) > “Threshold cointegration analysis of crude-oil benchmarks”. The Energy Journal 29/04, pp. 79-95, International Association for Energy Economics, and an e-mail from Thompson of September 14, 2008.]
 Because of time lags in processing and transportation, the oil industry makes greater use of forward and futures markets for hedging and speculative purposes than most industries. This has two major consequences —
[a] Oil markets are especially vulnerable to “players” in other markets who “roil” these markets by “laying off bets” made in oher markets, as when speculators in interest rates suddenly enter the Brent forward or futures markets.
[b] Oil markets are especially vulnerable to conflicts between different groups of speculators. We give two examples —
During the recent run up of crude prices to $147 per barrel, speculators in futures markets are estimated by some to have added $10 to $20 to oil prices. However, physical inventories seldom got “out of line” during this whole period.
One surmises that firms in the oil and finance industries with sufficient funds to rent and fill one or more tankers did not trust the duration of the run and were afraid to “get stuck” with large inventories when the run up collapsed, as it eventually did.
During the run down of 2009, obviously most speculators in futures markets were selling short. However, months before the bottom was reached, quite a few oil firms and three “Wall Street refiners” [investment houses] with “deep pockets”, rented and filled tankers at a cost of say $70 million each, in hopes of higher prices in the fall and winter. These in fact did materialize in due course, and most of these positions were liquidated at a profit, “unwound” in the jargon of the trade.
In stock markets, where most people play with much smaller stakes, such “bottom fishing” in anticipation of a low is frowned on as not only foolish but dangerous to ones financial health. So the fact that it did occur in both crude and product markets and on a much larger scale, indicates the degree of “speculative fever” among those who rented tankers.
 Dangerous bottlenecks for marine transportation.
For example, 20% of the world’s crude exports move through the Straights of Hormuz, between Iran and Oman. In the middle is an island full of Iranian Revolutionary Guards with guided. Most of Europe’s crude imports enter the Red Sea between Somalia and Yemen where Somali pirates have hijacked tankers, and al Qaeda is active in both countries. Another potential bottleneck where pirates have also attempted to hijack tankers is the two-mile wide strait between Indonesia and Singapore.
 Oil markets are normally fragmented to a considerable degree, by material and geography. However, it takes only two minutes for some “market rattling” news to spread around the world, and only ten seconds more for initial trades to be executed. So these fragmented markets can coalesce into one world market, almost before one can say “peak oil” !
As a result of the foregoing, crude-oil markets more often resemble a pack of rowdy hound dogs chasing each others’ tails than they do the stable, equilibrium-seeking markets of economic theory. And their future behavior is hard to predict.