Peak demand and oil's long-term trap
Fixating on the timing of a peak in oil demand is misplaced. Rather, the peak's significance is in shifting the paradigm, from perceived scarcity to perceived abundance. And it poses a problem for low-cost producers
The economic response would be to increase their output to capture value while they still could, pressing the advantage of cheaply extracted oil. But oil also carries a social cost in these countries. And the long-term oil price will therefore reflect not their marginal cost but the social one.
That's the subtle argument of a new paper from the Oxford Institute for Energy Studies, written by Spencer Dale, BP's chief economist, and Bassam Fattouh, the institute's director.
It's a compelling read. The notion of peak oil demand has been ubiquitous in the energy world's discourse over the past couple of years. The idea has lingered on during 2017, even as demand soared. If the oil-price rise of the past few weeks lasts through 2018 and starts to crimp demand growth, the peak demand argument will only get noisier again.
But the most significant part of Dale and Fattouh's article is surely the notion of the "social barrel". This neatly debunks the idea that oil is an ordinary market operating by ordinary rules. Saudi Arabia and other cheap producers may own the lowest-cost reserves in the world, but this is now irrelevant. These sclerotic oil-revenue-dependent states need a much higher price to maintain social order in their undiversified economies.
Thus a barrel might be taken from the ground for less than $10 in the Gulf, but it must sell for multiples more to fund a bloated social budget—maybe even to pay a subsidy for citizens to buy that barrel's refined product at a discount.
This will last as long as these economies have no alternative form of revenue to match oil. As Dale and Fattouh point out, such a transition won't happen overnight. Nor is it feasible to run a fiscal deficit indefinitely to pay for high spending during an oil-revenue drought.
All of which might sound like good news to higher-marginal-cost producers in diversified economies who don't need oil to pay for social welfare or subsidies. After all, demand peak or not, the world will still need a lot of oil: even if tight oil supply doubles or quadruples, to 10m or 20m barrels a day, write Dale and Fattouh, the world will still need 70m-80m b/d for the next "20 or 30 years".
And it should fetch a decent return, if the authors are right. The implication of their argument is that oil's price will depend not on physical lifting costs but on what Saudi Arabia and Opec producers need to balance their fiscal budgets.
Mind the demand gap
But don't get too comfortable. If this argument is correct, a trap lies in store for oil. If oil-demand growth begins to slow and even turn negative, yet prices are sustained well above marginal costs (to keep the Saudi and other budgets afloat), demand erosion could accelerate. Such elasticity is increasingly plausible in a world where new and genuine alternatives to oil in transport—its main source of demand—are emerging.
Functioning producers in functioning markets don't respond to waning demand for their product, or the arrival of alternative products, by keeping its price high.
On the other hand, as the authors point out, allowing the price to drop to the low-cost producer's marginal cost—with the resulting drop in social spending—would bring social problems. "At some point, the potential for ensuing socio-economic tensions could start to disrupt oil production and so cause oil prices to rise."
The only way out of this dilemma is for countries that can extract oil cheaply to learn to live within their means—and therefore use their low-cost oil strategically, capturing more market share and stimulating demand with a cheaper product. Yet this would need economic adjustment and upstream-capacity building that would take decades. In the meantime, prices in a distorted market will continue to reflect the higher-cost social barrel, not the lower-cost marginal one—even if the customer base starts to shrink.