Related Articles
Forward article link
Share PDF with colleagues

Missing barrels and data revisions

The International Energy Agency has changed its view of the oil market. What’s going on?

FROM “drowning in oversupply” to “light at the end of the tunnel”, the International Energy Agency (IEA) has in just three months accomplished what can only be described as a U-turn on its message to the oil market.

In its January monthly Oil-Market Report, the agency told readers the prospect of 0.6m barrels a day (b/d) of new Iranian oil output by the middle of the year could overwhelm the market, especially if Opec members maintained their output levels.

Already buckling under the weight of a 1bn-barrel build in OECD crude stocks between 2014 and 2015, oil inventories would add another 285m barrels this year, it said in the same report. Global oil supply would exceed demand by 1.5m b/d in the first half of 2016.

Then, in March’s Oil-Market Report, came a wholly different outlook. The IEA asked if there was “light at the end of the tunnel”, saying “there are signs that prices might have bottomed out”. OECD crude stocks may have drawn in February for the first time in a year, it added. 

The problem lies with the IEA’s frequent data revisions. And behind that are simply unreliable oil data.

Several recent examples stand out. In its February report, the IEA lowered its 2015 estimate for global crude demand by 100,000 b/d from the January figure, down to 94.4m b/d. In March, it increased that figure by 200,000 b/d and revised up its Q4 2015 estimate by 400,000 b/d.

Between January and March, the agency’s estimates of non-Opec crude supply for Q4 2015 rose by 300,000 b/d, to 58.1m b/d. But in its March report, it lowered its non-Opec supply forecasts for the second, third and fourth quarter of this year (compared with the previous month). For the whole of 2016, the IEA lowered the amount it expects non-Opec to produce by 100,000 b/d. And it’s not just in the monthly reports that the data tend to change. In its latest Medium Term Oil Market Report (MTOMR), released in March, the agency increased its global oil demand estimate for 2020 by 1.3m b/d above the estimates made last year.

Losing count


IEA data changes

The explanation was that it had underestimated European consumption in 2015 by 300,000 b/d; and Chinese crude demand was 0.6m b/d higher than it originally thought.

It all leaves market observers scratching their heads. Bullish ones are especially doubtful of the IEA’s data. Some question, for example, the agency’s basic calculation of the size of the glut – it says global crude production exceeded consumption by an average of 0.9m b/d in 2014 and 2m b/d in 2015.

The IEA records so-called missing barrels – the difference between production, consumption and reported stock changes – as data in the “miscellaneous to balance” line of its monthly reports. In the market, this is famous – or infamous, depending on your view – for triggering debates about the opacity of oil-market data.

In February’s monthly Oil-Market Report, the IEA showed a miscellaneous balance of 0.5m b/d for 2014 and 1m b/d for 2015. Maybe that oil was actually consumed, implying demand was much higher than thought. Or maybe it was never produced, and remains in the ground.

The most-cited explanations for the data discrepancies is poorly reported non-OECD demand and Chinese stock building. Taking the former line, Norwegian bank DNB Nor said in a recent report that crude demand in non-OECD nations may be “meaningfully larger” than the 48.93m b/d in Q4 2015 most recently estimated by the IEA. If so, the market is tighter now than the IEA thinks.

“The current oversupply is probably much lower than the 2m b/d reported by the IEA,” DNB Nor said. It went so far as to say the excess was probably half that number – and, it added, therefore the rebalancing would take half as long as the consensus suggests.

In its March report, the IEA had an answer, saying China may have accounted for more than half of the 1.1m b/d build in non-OECD crude stocks in the fourth quarter of 2015. It also lowered its missing-barrels figure by 200,000 b/d from the February report, down to 0.8m b/d for the whole of 2015. This was “well within the normal range considering the vagaries of oil data”, it wrote.

Seth Kleinman, global head of energy strategy at Citi, says IEA oil data should be taken with a pinch of salt. The adjusted numbers since the start of the year are a response to higher oil prices and frequent changes reflect “the enormous uncertainty about supply and demand” data.

“There’s a lot of confusion. Everyone is trying to read the tea leaves,” Kleinman says. “Their base case isn’t crazy. It’s a Herculean task at the best of times. But the problem is that people look to them for global supply and demand forecasts to justify price moves.”

Solving the mystery of the missing barrels, if they exist at all, is crucial to determining how quickly the market will rebalance and when – or if – prices will rise. Getting global oil supply and demand data right will also be essential.

The IEA remains the most comprehensive guide to a market in which too many important details remain far too obscure. Until the world’s oil market can produce and deliver perfect data, the agency’s numbers and predictions are going to keep fluctuating.

Also in this section
Gas breaks the oil price link
10 July 2020
The proportion of LNG priced on an oil-indexed basis fell to its lowest ever level in 2019
Letter from Houston: Energy hotspot hopes for a ‘bronze lining’
30 June 2020
The US oil capital may be battling both the oil price crash and a Covid-19 resurgence, but there are still reasons to be optimistic
Crude edges higher on Opec+ compliance promises
19 June 2020
Traders see the positives as producers commit to even greater respect for quotas