Should OPEC Now Raise its Output Quotas?
A. The Oil Price Saga
How far is Tipperary? Well, the answer to this question depends first and foremost on where you happen to be when you enquire. In the same vein we can ask: `How high are oil prices today?’ And we would then answer the question by a further question: `What is your reference point?’
Oil prices hit very low levels on Tuesday 9th February 1999. At the end of that day dated Brent was assessed at $9.70/barrel and WTI on the NYMEX closed at $11.68/barrel. The downward trend in prices which began in 1997 reached a nadir on that dark day. On Wednesday 11 August, almost exactly six months later, dated Brent was assessed at $20.57/barrel and WTI on the NYMEX closed at $21.52/barrel. The dated Brent price more than doubled and the WTI price rose by about 85 per cent!
The price rise appears to be huge if the reference point is the very low level where the decline bottomed out. A different picture emerges, however, if we compare current prices with the levels that have tended to prevail, save in three instances, throughout a recent ten-year period (1988-97). The exceptions were:
(a) The Iraq-Kuwait war. It caused prices to rise to high levels between August 1990 and mid-January 1991. In retrospect, considering the significance of this political event, the considerable damage caused to oil production facilities in Kuwait, and the embargo on Iraq the period of high oil prices appears to have been surprisingly short.
(b) A downturn in 1994. Oil prices fell down to $13 or $14 per barrel in reaction to the emergence of excess supplies. Compared with what happened later, in 1998 and early 1999, the oil price decline of 1994 now appears as a relatively short episode of limited significance.
(c) An upturn in 1996. During that year prices reached the $25/barrel level. The rise was generally attributed to low inventories. There is indeed an observed statistical relationship between crude oil prices on the NYMEX and oil inventory levels in PADD 2 (the US Mid-West) and PADD 3 (the US Gulf Coast). Prices started to fall away from this $25/barrel peak when inventory levels regained their normal levels.
One could broadly state that for the rest of the 1988-1997 period Brent prices have tended to average $17-18/barrel and WTI prices $19-20/barrel. For some reasons these price levels are accepted as `normal’ by the majority of oil producers (countries and companies) and by powerful governments of OECD countries. That they have been falling at about 2 or 3 per cent per year in real terms does not seem to have raised much concern. There is no doubt that the petroleum markets implicitly use these price ranges as references against which to assess the significance of price movements. Whenever the deviation from these reference levels appear to be unjustifiably large a correction sets in.
In the fifteen months period, from January 1998 to March 1999, oil prices deviated significantly from the usual norms. They moved, and stayed for a very long time, well below the levels that have come to be generally considered as acceptable, although not necessarily comfortable.
Markets marked down prices during this period: first, because oil production increased faster than demand, the latter being adversely hit by the economic crisis in Asia; secondly, because a contango on futures markets led to a build up of inventories which was then correctly interpreted as a supply overhang casting a long shadow on the months ahead; and thirdly, because in the period between June 1998 and early March 1999 the market lost faith in the ability of exporting countries to decide upon, and implement, a credible programme of production cuts.
In March 1999 the exporting countries got their act together. They agreed a package of output cuts to be implemented beginning in April which the markets judged to be credible for five important reasons.
(a) The size of the cuts, in excess of 2 mb/d was significant.
(b) Saudi Arabia agreed for the first time since 1993 to set its production well below 8 mb/d, a level that was considered for the past six years as immutable irrespective of the supply/demand balance on the market.
(c) The long-standing disagreements between Venezuela and the rest of OPEC were resolved with the election of President Chavez.
(d) The disagreement between Saudi Arabia and Iran about the Iranian quota and actual production levels which had paralysed OPEC during long months was removed by negotiations between the two countries conducted by their respective ministers of foreign affairs.
(e) The decision to curtail output was taken at a time when the supply/demand equation was roughly in balance, that is at a time when there was no further build up of inventories. This simply means that production cuts if implemented would, sooner or later, begin to reduce the supply overhang which the build-up of stocks in late 1997 and in the first eight or nine months of 1998 had created.
Markets always react in anticipation. They began to bid up oil prices towards the end of March, before the date at which the cuts began to be made. Because of inevitable lags due to the time it takes to bring oil to its destination, and to the fact that the first three months (April to June 1999) of implementation coincided with the seasonal fall in demand that usually occurs in the spring, inventory levels did not begin to fall before late in June, and more markedly in July.
The fall in inventory levels gave a new impetus to the oil price rise. We should recall, however, that prices move up or down, in response to news, anticipations or shocks from whichever level they happen to be at. This is precisely why commodity markets, like equity, bond on foreign-exchange markets, tend to over- or under-shoot on occasions. Anticipations had already pushed prices up when the decline in inventory levels send a new bullish signal which sent prices higher up.
B. Oil Prices are neither the Sole Criterion nor the Sole Objective
The critical question, now, is whether the exporting countries have reached, or even over-reached, their objective? Many voices are heard prompting exporting countries to reconsider their production policies, advising that quotas should be revised upward at the OPEC meeting next September, if not before.
Haven’t prices doubled in the past six months? Isn’t this a very dramatic increase, a significant shock that calls for a remedy?
Alas, these questions and views reflect a deplorable lack of understanding of the state of the market, the ways in which it behaves and the objectives of exporting countries. Once more, as in 1996 when prices reached temporarily a high level and in 1998 or early 1999 when they fell to abysmal levels, commentators who ought to know better want us believe that the high or low prices which obtain in a particular episode are there to stay for ever.
It is misleading to assess the oil price movement in relation to the trough of February 1999. The relevant observation is not that the oil price has doubled over the past six months but that after a very long interval of eighteen months or more it has returned to the level which most parties involved in the petroleum scene consider as acceptable or normal. The question that should concern the exporting countries, and indeed all those with a professional interest in petroleum, is whether all the fundamental features of the market reflect a return to a stable state?
The factors that caused the price collapse of 1998, apart from the inability of exporting countries to agree and implement swiftly a production policy were: production increases in the OPEC region in excess of the call on OPEC, that is from the difference between the increase in world demand for oil and the increase in non-OPEC production; the build-up of inventories which was helped by the emergence of a contango in the term structure of prices on the futures markets.
The current situation may be characterised as follows. Although oil prices have been racing ahead inventory levels are not yet back to the levels that ensure some market stability and the term structure of futures prices is not yet entirely in backwardation.
To assess these features of the current market we need a point of reference for comparison. Neither the 1996 nor the 1998 market situations provide valid references because the market was in disequilibrium in both these years. A better, though not ideal, reference period may be the first half of 1997. During these six months prices moved back from the high levels attained at the end of 1996 to the normal ranges of $17-18/barrel of Brent and $19-20/barrel of WTI; inventories were built up from low levels to more usual ones, and the steep backwardation on futures markets gradually became flatter and flatter. The averages of relevant data for the six first months of 1997 (an approximation of the period’s mid-point) provide a reasonably good idea of the features that characterise a market in a state consistent with the desired price ranges.
The average levels of crude oil inventories in the first half of 1997 in the USA and in PADD 2 and PADD 3 were as follows:
USA PADD 2 PADD 3
309.2 mb 67.8 mb 150.3 mb
On 30 July 1999 the respective levels were as follows:
USA PADD 2 PADD 3
325.6 mb 71.5 mb 167.8 mb
We are considering these particular data instead of estimates of world-wide inventories not only because US statistics are more reliable but because the WTI price is very sensitive to oil inventory levels in the USA and particularly in PADD 2 and PADD 3. It is clear that the inventory levels are still higher than their comfortable levels.
Although the WTI price on the NYMEX reached $21.52/barrel on 11 August 1999, the term structure of futures prices showed on that day a small contango between the September contract ($21.52/barrel) and the October contract ($21.61/barrel). Backwardation only appears as from the third month. In 1997 the term structure shifted from steep backwardation at the beginning of the year (68 cents between first and second month and $3.19 between first and sixth month on 1 January 1997) to a flat price line in March and a mild contango of about 10 cents per month in June.This inversion was the first symptom, unnoticed at the time, of the crisis to come. It is clear that the term structure of futures prices at the beginning of this August does not clearly suggest that supply and demand are in stable balance, and certainly not that the market is tight.
C. Conclusions
We conclude that the time has not yet come for exporting countries to decide on an upward revision of their quotas. They may all draw great comfort from the rise on oil prices. They should remember, however, that prices do over- or under-shoot. They often move ahead of other indicators. The oil-exporting aim was, and still is, to remove the supply overhang which is still there in the form of surplus inventories. Oil prices would become vulnerable to a strong downward correction if this aim is not achieved. And they will inevitably fall in a significant manner if premature increases in production leads to a new inventory build-up, to the re-emergence of a contango on futures market, and to a change of the market’s perceptions of the motives and behaviour of oil-exporting countries.
