Mar 3rd 2011 | BREGA, LONDON AND WASHINGTON, DC | from the print edition of THE ECONOMIST
TWO factors determine the price of a barrel of oil: the fundamental laws of supply and demand, and naked fear. Both are being tested by the violence that is tearing through Libya, the world’s 13th-largest oil exporter. The price of a barrel of Brent crude now hovers around $115. On February 24th, however, it rose to almost $120, as traders realised that they might have to do for a while without some or all of Libya’s exports: some 1.4m barrels a day (b/d), or about 2% of the world’s needs.
The situation in Libya is grim, as the rebels and the forces of Muammar Qaddafi battle for control of the country’s only resource. Brega, the seat of the Sirte Oil Company in the east of the country, has changed hands three times in recent days. Most of the oil workers have fled, and production has fallen by two-thirds. The ports of As Sidra, Brega, Ras Lanuf, Tobruk and Zuetina, which together handle almost 80% of Libya’s oil exports, were all seized by the rebels; two have now been retaken by Colonel Qaddafi’s forces. The rebels remain in control of Africa’s largest oilfield, Sarir, pumping some 400,000 barrels on a normal day. But for how long?
The history of oil is marked by Middle Eastern strife, supply shocks and global recession, with the Arab oil embargo in 1972, the Iranian revolution in 1978 and Saddam Hussein’s invasion of Kuwait in 1990. To gauge the risks today you need to answer three questions. How vulnerable is the oil market to an interruption in supply? How sensitive is the world economy to oil-price spikes? And how well can policymakers cope with a shock if the worst happens? Take each in turn.
The troubles in Libya are only the most serious example of the impact of Arab unrest on global oil markets. Prices jumped as Egypt’s citizens took to the streets to oust President Hosni Mubarak. Egypt is an oil importer, but acts as a vital conduit between the huge oilfields in the Persian Gulf and markets in Europe, via the Suez Canal and through the SUMED pipeline. Although it seemed unlikely that protesters would or could disrupt oil shipments, events in Cairo were enough to add more than $5 to a barrel.
The spread of unrest to Bahrain, Oman and the Gulf has created a whole new dimension of anxiety. North Africa produces 5% of the world’s oil, but the Middle East produces 30%. Moreover, Bahrain’s problems are on Saudi Arabia’s doorstep. These bear on the situation in the eastern Saudi provinces, from which a huge quantity of oil is pumped into global markets.
Saudi Arabia is therefore the traders’ chief worry. But it is also, in oil terms, the world’s chief hope. It is the only producer with significant spare capacity that could quickly be released if the oil price rose too high. Although OPEC, in which Saudi Arabia is the biggest force, exists to keep oil prices buoyant, it does not want to see them reach a point where the world economy is damaged and demand for oil falls. When prices spiked in 2008, the Saudis said they had capacity to spare. Terrified oil markets doubted its existence, and prices rose anyway, to reach $145. Yet the subsequent collapse in the oil price in the second half of 2008 was only partly caused by the credit crisis and the rich-world recession that resulted. Saudi Arabia also pumped extra oil: nearly 2.5m b/d on top of the 8.5m it was already providing.
OPEC’s spare capacity now is put at anything between 6m b/d (by OPEC) and 4m-5m b/d (by industry analysts); Saudi Arabia’s share of that excess is perhaps 3m-3.5m b/d. The oil price has retreated from its peak in the past ten days largely because Saudi Arabia says it is pumping up to 600,000 b/d to replace the shortfall in Libyan exports. It has invested heavily in expanding capacity, with plans to spend perhaps $100 billion on wells and infrastructure by 2015. It has also been far more open about letting the world see what it has done. OPEC’s stated aim of stabilising oil prices relies on traders believing that the Saudis really do have the capacity to pump more when prices rise.
Why, then, are traders still so nervous? The answer is that the long-term trends of supply and demand were already unfavourable when the Arab shoe-throwers intervened. Before the uprisings, a barrel of Brent crude was commanding close to $100 a barrel. World demand grew by an extraordinary 2.7m b/d in 2010, according to the International Energy Agency. It will probably keep growing by another 1.5m b/d this year and the same again next, as the rich world recovers and demand surges in China and the rest of Asia. Net expansion of non-OPEC supplies is likely to be negligible in the coming years. Though the rich world’s inventories are high, with cover of around 50 days, it is not clear that Saudi Arabia can pump much more than it did in 2008; and the speed of oil released from government reserves, such as America’s Strategic Petroleum Reserve, also has upper limits.
If disturbances hit Algeria and threaten its oil industry too, the buffer of spare capacity would fall below where it stood in 2008. But demand now is much higher, so spare capacity as a proportion of that demand is much lower (see chart 1).
When oil markets tighten, another set of problems emerges. Saudi oil is generally more dense and sulphurous than the Libyan crude it will replace. Europe’s creaky old refineries will not be able to process the heavier Saudi crude, and fuel regulations there are less tolerant of sulphur content than elsewhere in the world. So the Gulf oil will have to be shipped to Asia’s newer refineries, which are designed to deal with a wide variety of grades of oil. West African oil, a close substitute for Libya’s output which usually goes to Asia, will be sent to Europe instead.
If the supply situation worsens, opportunities for this type of substitution will be fewer, creating supply bottlenecks, shortages of petrol and spikes within price spikes for different crudes and products, even when spare capacity remains. The price differential of about $15 a barrel that has built up between Brent crude, which more closely reflects global trade, and West Texas Intermediate, the benchmark for oil prices in America, is a good example of how oil markets can become distorted by local patterns of supply and demand. If supply gets even more stretched, oil could fetch a far higher price in some parts of the world than others. If supply problems become really grave, oil companies may even declare force majeure, raising the prospect that, as in 1978, oil markets fail altogether.
That is still a remote prospect, and the upward march of the oil price seems to have paused for now. The crucial question is how much oil will be lost, and for how long. When oil markets operate at the limits of supply, even the smallest extra disruption has a disproportionate effect. On February 26th, for example, Iraq’s biggest refinery shut down after a terrorist attack. This and other assaults could knock out another 500,000 b/d from the world’s fuel supplies. And if the raids on oil installations in previous elections in Nigeria are anything to go by, the next one, in April, may threaten another 1m b/d of supplies from west Africa. Meanwhile, Saudi Arabia remains far from secure (see article). On March 1st the country’s stockmarket, jittery about the neighbours, plunged by 7%, a worrying sign that confidence is fading.
Hobbling the world
All this is a dark cloud on an otherwise bright horizon for the global economy. Few things can short-circuit growth like an oil shock, both because of the fuel’s ubiquity and because of the relative insensitivity of demand. When the oil price jumps, consumers have little choice but to accept it, spending less on something else.
So how sensitive is the world economy to oil prices? Thus far the rise, and the likely damage, both look modest, in part because many forecasters had expected an increase this year anyway. Since the end of last year the price of Brent has risen by $23 a barrel, or about 25%, and West Texas Intermediate by $10, or 10%. The IMF reckons that a 10% increase in the price of crude shaves 0.2%-0.3% off global GDP in one year. As it happens, crude oil (using a blend of several grades), is now about 10% more costly than the IMF assumed in late January, when it projected global growth of 4.4% this year. That implies that the Fund would now foresee growth of about 4.2%.
Economists do not expect a repeat of the 1970s, when oil-price rises led to “stagflation” in the rich world. Olivier Blanchard, chief economist of the IMF, and Jordi Galí, of the Centre de Recerca en Economia Internacional in Barcelona, point out that two recent oil-price rises—one beginning in 1999 and another in 2002—were of the same order of magnitude as during those turbulent years. But the effect on both inflation and unemployment in the rich world was much smaller: in America, for example, a rise in inflation of only 0.7 percentage points on average, whereas the 1970s shocks had caused a rise of 4.5 points in the two years after the shocks.
The rich world is less vulnerable now because it has substantially reduced the amount of oil used per unit of output. America’s economy in 2009 was more than twice as large in real terms as in 1980. Yet over that period America’s oil consumption rose only slightly, from 17.4m b/d to 17.8m. Europe actually used less oil in 2009 than in 1980, even though its economy had grown.
Other factors may also have helped. Supply shocks generate larger increases in inflation and bigger falls in output when wages are rigid. So oil shocks have smaller effects today, because labour markets in rich countries have become considerably more flexible since the 1970s.
Emerging economies may be hit harder by a spike, since they use more oil per unit of output than rich countries do (see chart 2). America’s economy, though about three times the size of China’s, uses just over twice the amount of oil that China’s does. But oil intensity in emerging countries has also been falling in recent years, as manufacturing has become more efficient and less energy-intensive service industries have increased their share of the economy. So even these countries are less vulnerable to an oil shock than they used to be.
Among rich economies America tends to suffer the biggest immediate impact, because its economy is relatively energy-intensive and because its low petrol taxes interpose only a small wedge between crude oil and petrol prices. Goldman Sachs estimates that a 10% price increase trims GDP by 0.2% after one year, and 0.4% after two.
In Europe the effect is muted by lower energy intensity and high levels of tax. Excise and value-added tax represents roughly 60% of petrol prices and 52% of diesel prices in the euro area, according to the European Central Bank (ECB).
Emerging Asia is more complicated. Although its economies are more oil-intensive, several also export oil, and many subsidise fuel, limiting the impact on consumers. Thailand has resolved to hold the price of diesel below 30 baht (about $1) a litre until April; without the subsidy, which was raised on February 24th, it would be 34 baht. Citigroup estimates that each $10 increase in the price of oil costs India’s state-owned oil-marketing companies the equivalent of 0.5% of GDP, of which half is absorbed by the budget. An IMF staff study has estimated that emerging and developing countries subsidised fuel by about $250 billion in 2010.
Loosening, or tightening?
What can central banks do to protect the economy? Higher oil prices act as a tax on countries that import the stuff, which would normally call for easier monetary policy. But they also raise inflation, which calls for tightening. A one-off rise in prices would not produce a sustained increase in inflation, unless it boosts firms’ and workers’ expectations of future inflation, which can become self-fulfilling. The oil-price shocks of the 1970s rapidly found their way into broader inflation. Central banks had to clamp down drastically to suppress their inflationary effects.
In recent years, with inflation expectations more stable, central banks have responded more moderately to higher oil prices. But in July 2008 the ECB raised short-term interest rates because it feared that a rise in headline inflation would feed a wage-price spiral. In retrospect, that was a mistake. The global economy was already slowing, and over the next year both headline and core inflation (which excludes energy) fell sharply in the euro zone. Although America’s Federal Reserve did not tighten, it hinted at the possibility, which prompted markets (wrongly) to anticipate a rate increase. These hawkish signals may have compounded the slide in economic activity already under way.
This year their response is likely to be more subdued. Unemployment is higher in America and Europe than in 2008, and underlying inflation, except in Britain, is lower. At a forum on February 25th at the University of Chicago, officials from both the European and American central banks signalled willingness to hold fire unless inflation expectations grow. On March 1st Ben Bernanke, chairman of the Federal Reserve, said the recent rise in commodity prices would probably “lead to, at most, a temporary and relatively modest increase” in inflation.
In many emerging markets the risks are greater. Those economies are already operating at capacity, and both overall inflation and core inflation have risen: China’s January inflation rate was 4.9%, well above its official 4% target, and India’s was more than 9%. An increase in the price of energy can cause a steeper jump in inflation in emerging markets, because in many it has a larger weighting in their consumers’ baskets: 15.2% in Indonesia, 14.2% in India and 13.8% in Malaysia, compared with about 9% in America’s. Moreover, energy is a large input in food production, which has an even bigger weight.
Monetary policy has also been relatively loose in these countries, with real short-term interest rates negative in many of them, including, by some measures, China. Johanna Chua of Citigroup reckons that monetary conditions, including both interest rates and the exchange rate and, in China’s case, credit growth, have tightened already in Asia, but need to tighten further in both China and India.
The reason for a rise in the oil price is as important as how large it is. An increase forced by higher demand is less dangerous than one driven by constricted supply, because it is evidence of a healthy global economy. If rapid growth means that China and India are importing more oil, they are probably importing larger amounts of other things as well, lessening the pain for slower-growing consumers of oil.
Nonetheless, whether driven by demand or supply, a large enough spike in the price of oil can do great damage. Economists call such abrupt responses “non-linearities” and they suggest that when the price rises fast enough, consumers and businesses trim their spending and investment plans. This is often because prices are driven by other factors that hurt confidence, such as wide unrest in the Middle East. If another Arab government were toppled, pushing the oil price over $150, the economic impact would almost certainly be larger than the 0.5% to 1% of GDP that simple extrapolation suggests.
James Hamilton, of the University of California, San Diego, has identified numerous periods since the late 19th century in America when an abrupt rise in the price of oil or petrol coincided with recession. Many of these were caused not by an interrupted supply, but by demand growth colliding with unresponsive supply. That seems to explain the price spike above $140 in mid-2008. Although the financial crisis was the main cause of the recent recession, Mr Hamilton argues that oil explains why the economy had already begun contracting before the worst of the crisis hit that autumn. Robert McNally, of Rapidan Group, a consultancy, concurs, arguing that American consumer confidence fell sharply once petrol went past $3 a gallon (see chart 3). It is now at $3.38, after the biggest one-week increase since Hurricane Katrina in 2005.
Even if the unrest leaves supply unaffected, significantly higher prices may be only a matter of time. The same dynamic that drove the oil price skyward in 2008 is steadily reasserting itself. Supply is not growing substantially, and global demand, which regained its pre-recession peak last year, is expanding briskly again.
Given enough time, the rich countries should be able to adjust to higher prices. Jim Burkhard of IHS Cera, a consultancy, notes that OECD oil demand peaked in 2005 and has been slipping since in response to the upward march of prices. In America a shift in consumer purchases towards more fuel-efficient vehicles, ethanol mandates and higher fuel-economy standards have all capped growth in petrol demand. Meanwhile, the higher world price has unlocked new supply within the United States, and elsewhere, which was previously too expensive to exploit.
Yet it may take years for such trends to dent demand and boost supply by much; and the world may not have a lot of time. “Historians will look back on 2008 as the first time in modern memory that spare capacity ran out without a war in the Persian Gulf, and OPEC failed to cap prices,” says Mr McNally. “Eventually we’ll replay that scenario. If OPEC can’t control the market any more, that means prices will have to swing much more.”