Does Oil Price Volatility Matter?
There are different kinds of oil price movements. The first case, which today is only of historical interest, arises when oil prices in international trade were administered first by a group of major oil companies and subsequently by OPEC. The oil price behaviour followed then a typical pattern characterised by periods of different lengths during which a reference price (eg Arabian Light 34 API) remained constant in nominal dollars. We had therefore price episodes, and the passage from one episode to the other was the result of a punctual decision by the entity which happened at the time to administer the price of oil. We had then occasional price shocks rather than continuous volatility. The fundamental difference is in the nature of the adjustment process that shocks and volatility induce.
The second case describes a price regime that was introduced in 1987 and that is still ruling today. OPEC no longer fixes the reference price. The exporting countries now sell oil in international markets on the basis of price formulae which use as reference the spot or futures prices of certain marker crudes, namely WTI, Brent or Dubai. The behaviour of prices in the world petroleum market is essentially that of these marker crudes. Volatility therefore arises in the complex and interrelated set of spot, futures and other derivatives markets.
Volatility is simply a characterisation of price changes over time. In futures markets the changes are almost continuous. They occur both within and between trading days. Prices change in responses to ‘news’ – that is to a very wide variety of information data which influence traders’ views on whether it is opportune to buy or sell. If the news, for example, are bullish some will want to buy and they will raise the bid price to the level that will persuade others to sell.
Relevant news do not exclusively relate to the exact state of the supply/demand balance which in any case is generally unknown. Traders also respond to news that alter perceptions of future market developments such as policy statements, economic forecast, industrial events etc. As important are the information that a trader may gather or the guesses s(he) may make about the positions taken by other traders and their trading optimization strategies.
Trading requires volatility. Without it there will be no need to hedge and where there are no hedgers, there are no speculators. But trading can also cause additional volatility; additional here refers to price changes brought about by strategies which are not determined by responses to perceived changes in current or future fundamentals but by the search for pure trading profits.
The many causes of volatility can be listed, at least theoretically. To disentangle them in an attempt to attribute their relative contribution to a sequence of price changes in an empirical study may prove impossible. But these difficulties should not deter policy makers who may be concerned by the adverse impact of price volatility on economic behaviour of tackling factors known to unsettle markets or to cause prices to over- or under-shoot.
There is no doubt that significant improvements in the quality, reliability and accessibility of the relevant information, other things being equal, reduce the degree of volatility. Measures that limit the ability of punters to mount squeezes would by definition have the same effect.
It is useful to distinguish short-term price fluctuations from episodic movements that sometime characterise certain longer periods of time. The most dramatic episode occurred fairly recently and is still very alive in people’s minds: this is the 1998/ early 1999 price collapse followed by rises which took prices to high levels throughout 2000. The WTI price (NYMEX first month futures contract) was at $17.65 per barrel at the beginning of January 1998. It reached a low of $10.80 in late December 1998, but the lowest levels were not hit until early February 1999 when WTI bottomed at $10.26 and Brent at $9.70. After that date the price movement was relentlessly upward with the WTI price ending the year at around &26.50 per barrel and peaking at $34.15 on 7 March 2000. It took 13 months of toil for the market to bring the price down by slightly less than $7.0 (that is by 39%) and then another 13 months of over-excitement to raise it by almost $24.0 (that is by 233%).
This is by far the most damaging kind of price movements. It is important to note here that the adverse impacts are not only due to the magnitude of the fluctuations but to the price levels attained at the extreme points of the range: the low $10 and the high $34 prices.
Three sets of causes are behind these violent movements.
First, the market’s perceptions of the fundamentals. In early 1998, the market reacted to the OPEC decision made in Jakarta in late November 97: to raise quota levels by 10%. I never believed that the Gulf producers (the only ones with surplus capacity) had actually increased their output by this amount, but that is largely irrelevant since what matters most in the short run is what the market believes, not what actually is.
Secondly, the market’s perceptions of the state of OPEC’s solidarity. In 1998, and until the advent of the Chavez administration in Venezuela in February 1999, the market had a dim view of OPEC’s ability to ‘get its act together’. And this negative view proved stronger than the positive impact that two successive cuts in production (March and June 1998) should have had.
Thirdly, the dynamics of futures markets. When the term price structure is in contango, and provided the positive price differential between two successive contract months is sufficiently large, there is an economic incentive to build up stocks. But a stock build is almost invariably considered to be a ‘bearish’ factor, indicating that oil is overproduced. We thus have a vicious circle where low prices cause the contango to become steeper, encouraging further stock build which lead to a further price fall. The vicious circle ceases to operate when storage fills up and the costs of any addition to stocks become prohibitive.
Similarly, when the price structure is in backwardation the incentive is to draw from stocks. This is usually interpreted as a ‘bullish’ signal, an indication of underproduction, which causes prices to be bid up. A vicious circle comes into operation until the industry begins to worry about the low stock levels reached.
Identifying these causes helps the search for remedies. It is clear, for example, that regular, authoritative and credible information on OPEC’s production would reduce the negative impact of the first cause listed above. There is no doubt also that OPEC member countries can remove the second major cause of instability by avoiding disagreements on minor issues and keeping negotiations undertaken to settle disagreements as secret as possible. The recent stability of oil prices owes very much to the solidarity displayed by OPEC member countries and restraint over the discussion of potential disagreements in public. Finally, a contango can be dealt with by refusing to meet demand that is clearly directed to stock building.
The rise that took the WTI price from $10 to$35 per barrel over a 12 months period was not entirely due to the producers’ unwillingness to supply crude oil. Other important causes were a) the steep backwardation which affected that part of the demand for oil relating to inventories and b) a shortage of products in the USA which caused their prices to rise. Crude oil price movements follow with lags the general tendency of product prices (and vice-versa).
The volatility of crude oil prices (and indeed any other characteristic of their behaviour) cannot be fully understood when the focus is put exclusively on the crude oil market and on OPEC’s policies. The downstream market and refiners’ behaviour also matter.
Besides the occasional big price movement, such as the 1998-March 2000 one, which are in the nature of a double shock, there are shorter term fluctuations which may pass unnoticed but are nevertheless very significant. Price movements of the order of $7-10 per barrel over short periods of 2-10 weeks occurred in several instances during 2000.
- Between 7th Jan and 7th Mar 2000 the WTI price rose by $9.50 per barrel
- Between 7th Mar and !0th Apr the WTI price fell by $10 00 per barrel
- Between !0th Apr and 20th Jun the WTI price rose again by $9.50 per barrel
- and Between 27th Nov and 14th Dec the WTI price fell again by $7.50 per barrel.
These are significant movements amounting sometimes to a 30% variation in a matter of few weeks.
Does any of that matter?
Ask any trader and the answer will be almost invariably ‘no’. The alleged economic rationale is that we are considering a market which sends signals about the allocation of resources. That would be perfectly correct if these price changes caused rapid adjustments in supply and demand. But they do not because the commodity traded is not physical oil but either a claim on future oil or a price differential which is no a commodity at all.
The signals that the industry needs are about investments in capacity. One could argue that the prices quoted for long term swaps provide these signals, and indeed long term prices fluctuate much less than short term ones. But investments do not depend only on the view taken about prices in the long term, they are strongly influenced by the cash flow available to companies and the cash flow is a function of current prices.
Volatility generates uncertainty and uncertainty inhibits or confuses the investors.
The big price movements of the 1998 type are disastrous because they end up by curtailing investment plans significantly. This forces service companies, on whom much depends, to lay off personnel with damaging losses of skills and experience.
Volatility disturbs governments of exporting countries as they rely heavily on oil revenues. Low prices lead to severe curtailment of expenditures, but such as the constraints of domestic politics that the axe does not always fall on the less worthy projects. High prices lead to demands for expenditure increases that are not sustainable in the long run.
Price instability generates instability on a wide front: investments, human capital, corporate performance and the economic development of oil exporting countries. That it serves trading interests is of no great consolation. And price instability feeds itself on itself because it induces OPEC into remedial courses of actions that backfire sometimes, given uncertainties about the forces at work, misinformation and faulty interpretation by the market of OPEC’s intentions.
Stability does not imply fixed prices. A certain amount of flexible variations is both necessary and beneficial. What is required is a market that signals correctly the state of the current and the expected future balance of the demand for and the supply of oil. There is clearly a need for a fundamental market reform. This will require the co-operation of all the major players. We are not yet there: the understanding of the issues leaves much to be desired and the political will is very weak. Sooner or later, however, the adverse effects of excessive volatility and damaging price shocks will induce a search for remedial action.