Brent futures prices tumbled more than 6 per cent on Monday, an unusual move that has raised questions about whether shifting fundamentals will send the market even lower over the coming months.
Front-month prices have dropped 17 per cent over the past month, breaking out of the very narrow range that had prevailed since the middle of April.
Reasons for the drop are not hard to find. The Financial Times on Tuesday set out the six bearish factors that have come together to produce a perfect storm on the oil market.
These include China’s stock market tumble, turmoil in Greece, prospects for a nuclear deal with Iran, rising oil output from Opec, an increase in rigs drilling for oil in the United States for the first time in more than six months and bearishness among commodity-focused hedge funds and banks.
But the significance of the big move on Monday and the wider decline in oil prices over the past month is much harder to assess because large moves in the price of oil and other commodities do not correlate with new information about supply and demand.
Of the six factors identified by the FT, we can discount three (Greece, Iran and Opec) as explanations for the big move this week and over the past month because they have long been known and should already have been incorporated into prices.
That leaves China, US rigs and increased bearishness among hedge funds and commodity banks as new information to explain the sudden plunge in prices.
None of them offers a particularly convincing explanation for such a large price shift in such a short time.
China’s equity market has lost almost a third of its value since the middle of last month and so far has resisted all attempts by the authorities to stabilise share prices.
There are fears that the slump will worsen the country’s economic slowdown and translate into slower demand growth for a broad range of commodities from crude and coal to iron ore and copper.
There is certainly some risk to the economy of the world’s largest oil importer, but the link between the stock market and commodity consumption remains tenuous.
“Political risks associated with the current fall (in share prices) are far greater than the economic risks,” said Michal Meidan of the consultancy China Matters on Wednesday.
“There is a clear disconnect between economic fundamentals and the stock market – this disconnect was just as palpable when the market was making dramatic gains as it is in the current bearish sentiment.”
Ms Meidan argues the “stock market crash will have limited impact on the domestic economy and oil-demand growth”.
US rigs rise
The other new information in recent days was the jump in rigs drilling for oil in the US last week.
The number of rigs increased by 12, the first rise since December 5 after 29 consecutive weeks in which the oil rig count declined.
It has been clear for some time that the downturn in drilling was bottoming out, but the increase in active rigs was a surprise.
It comes after some shale companies had begun to talk about the prospect of ramping up their drilling programmes and production in the next 18 months if prices continued to rise.
For bearish commentators, the climb served as confirmation that oil prices had risen too far, too fast and were not consistent with the need for a further slowdown or even fall in shale production.
“Not only did US$60 per barrel oil, strong high yield and equity energy markets create an increase in US drilling last week, but the market structure of the new oil order has generated incentives for low-cost producers such as core Opec and Russia to ramp up current and future production,” Goldman Sachs wrote.
“We reiterate our fundamentally driven forecast for lower oil prices,” Goldman concluded.
But there are reasons to be cautious. The increase in rigs was very small – just 12, or 0.2 per cent, after the rig count had fallen by 981, or 61 per cent, over the previous eight months.
Rigs counts show considerable variability (the standard deviation of the weekly change is 13 rigs), so the statistic is noisy and last week’s increase was not in itself significant.
Moreover, the rise in rigs drilling for oil was almost matched by a drop in the number of rigs drilling for natural gas. The nine-rig decline was the largest in 13 weeks.
It is possible that some rigs are being redirected from gas-rich to oil-rich formations or simply that drilling targets are being reclassified from gas to oil.
Some stabilisation or even increase in the number of rigs drilling for oil has long been anticipated – the decline could not continue indefinitely and the industry needs to replace declining production from existing wells.
But can such a marginal increase in the number of rigs drilling for oil explain a fall of 6 per cent in oil prices on a single day or 17 per cent over the last month?
Whatever the explanation, the decline in the front-month Brent futures contract on Monday was unusually large.
In the quarter-century since the start of 1990, prices have fallen by a greater percentage on only 54 days out of almost 6,500.
Large falls have sometimes coincided with fundamental news about supply and demand, such as when Opec decided in November last year not to cut its production despite the slide in prices.
However, large falls have also occurred in the absence of significant news about supply or demand, as happened during the flash crash in May 2011.
The relationship between news flow and short-term price moves is comparatively weak and makes linking the two notoriously hard.
Sentiment and market positioning often exert a bigger influence on prices in the short term, even if fundamentals are decisive in the medium and long term.
Sentiment and fundamentals need not be mutually exclusive explanations. Sometimes a shift in sentiment can crystallise a long-building reaction to fundamentals.
“The fall in oil prices (this week) is a story of fear and fundamentals,” the historian Daniel Yergin wrote. “Fear is the trigger for seeing fundamentals of supply and demand more clearly.”
But sometimes abrupt price shifts reveal nothing more than a crowded market, with too many participants trying to trade in the same direction at the same time.
Oil forecasters are vulnerable to availability bias
Why is there such good data about oil in the United States but such poor data about everywhere else?
Accurate information is essential for good decision-making, so it is remarkable how little reliable and timely data exists about the production and consumption of crude oil and refined fuels outside the United States.
The situation in the other economies, especially emerging markets, is mostly guesswork.
The result is that oil analysts cannot even agree on production and consumption yesterday and today, let alone predict what will happen tomorrow.
And because the best and most readily accessible data is for the United States, the market puts excessive emphasis on what happens there and neglects developments elsewhere.
The obsession with weekly rig counts, production estimates and crude inventories in the United States as a sign of wider supply-demand trends in the oil market has been a case in point.
But as long as US data is more accurate, detailed and timely than the numbers for other countries, this example of availability bias is set to continue.
Some US data comes from private companies such as Baker Hughes, which inherited the decades-old rig count from the Hughes Tool Company, but most is produced by the federal government.
The US energy information administration, the independent statistical and analysis arm of the Department of Energy, provides by far the best data on oil and other energy markets anywhere in the world.
The problem is that no other country provides anything like the same quality and depth of information about its energy industries.
Other advanced economies such as the UK, Germany and Australia provide data that is less comprehensive and far less timely, often on websites that are byzantine in their complexity.
In the case of emerging markets such as China and Saudi Arabia, the data is either missing or considered a state secret.
With emerging markets accounting for more than half of the global oil demand for the first time in 2013 and last year, most of the oil market is now opaque.
Inevitably, analysts, traders, investors and journalists tend to focus on the most readily available information and then extrapolate to the rest of the market, a variant on the data availability bias identified by behavioural economists.
Congress’ preoccupation with data in the 1970s explains why there is such rich information about oil in the US.
But it has created an enormous distortion. Better data on the rest of the world – especially emerging markets – must be the top priority if policymakers want markets to operate more smoothly.
Follow The National’s Business section on Twitter