Improved fuel efficiency and substitution of natural gas will restrain world oil demand over the next decade.  But will these effects be enough to offset increasing demand from emerging economies? This important debate could be sharpened by clearer economic language.

A conventional view of world oil demand is that it will continue to grow for many years  as countries such as China and India enjoy rapid growth, creating huge new demands for the transport of people and goods.  But this is challenged in a recent report by Citi Research entitled ‘Global Oil Demand Growth – The End Is Nigh’.

Based on research by its automobile equity team, Citi Research forecasts annual fuel efficiency improvements for new vehicles of 2.5% per annum as manufacturers respond to standards coming into effect in many countries and to high oil prices (1).  That may not seem a lot, but it implies a halving of fuel consumption per mile in just 28 years.  It also forecasts significant substitution of natural gas for oil in transport, petrochemicals and (in countries with oil-fired power stations) electricity generation (2).  In the US, substitution is already occurring due to cheap shale gas.  In other countries it is expected to accelerate from 2016 with the completion of many LNG (liquefied natural gas) export projects (3).  The report also shows that these trends apply to emerging as well as developed economies.  China, for example, has introduced vehicle fuel economy standards which will take effect in 2015, and is establishing an infrastructure network for refuelling natural gas vehicles (4).

Allowing for these factors, Citi Research projects world oil demand as plateauing at about 91 mb/d (million barrels per day) from 2018, only slightly above its 2012 level of 89 mb/d (5).  But here I have a problem.  I was brought up, economically speaking, to distinguish between a change in demand, meaning a shift of the whole demand curve, and a change in quantity demanded, meaning a movement along the demand curve associated with a change in price.  When I read a prediction that demand will be so many million barrels per day, I wonder what this means and what is being assumed.

Focussing on the oil demand curve and making the usual ceteris paribus assumption, I would say that improved fuel efficiency will shift the whole curve, reducing quantity demanded at any particular price.   Increased supply of natural gas and increased capacity to use it as a transport fuel will shift the curve in the same direction.  On the other hand, rising incomes in emerging economies will shift the curve in the opposite direction, increasing quantity demanded at any particular price, even if the cars bought by the growing middle classes and the vehicles that supply their goods are highly fuel-efficient.  So I was initially inclined to interpret the report as implying that these opposing forces affecting the demand curve are now almost in balance, and will reach a balance from around 2018.

But then I noticed that Citi Research also predict a fall in oil prices from their current $100 per barrel to $80 – 90 by 2020 (6).  A levelling-off of quantity demanded at a time when price is falling implies a fall in demand – a shift of the demand curve to the left (see diagram).  This assumes, as most studies suggest, that oil demand is moderately price-elastic (7).  If the price and quantity predictions are taken together, therefore, the implication is even more striking than the report’s title suggests: a fall in world oil demand over the period to 2020.  In other words, the forces tending to reduce oil demand will more than offset those tending to increase it.

This is perhaps to read too much into Citi Research’s report.  Unfortunately, it does not state whether the demand projection takes account of the projected price fall or is based on constant prices.  Moreover, the chart showing this projection was reproduced in a recent feature in The Economist (8), where it was no doubt much more widely seen, but remained subject to the same lack of clarity.

This debate over future oil demand is important, not least because substitution of natural gas for oil (and coal) is one of the most practicable ways of reducing carbon emissions and mitigating climate change – a point highlighted by Dieter Helm in his book The Carbon Crunch (9).  It is perhaps slightly unfair of me to single out Citi Research for a looseness in economic terminology of which many others have been guilty.  But it would certainly make for a sharper debate if participants would indicate clearly what they mean when they predict changes in demand.

Notes and References

1. Citi Research (2013) Global Oil Demand Growth – The End Is Nigh  p 2

2. Citi Research as above p 1

3. Citi Research as above p 5

4. Citi Research as above pp 6 & 9

5. Citi Research as above p 2 Figure 1

6. Citi Research as above p 1

7. Hamilton J D 2008) Understanding Crude Oil Prices p 34 Table 3  http://dss.ucsd.edu/~jhamilto/understand_oil.pdf

8.   The Economist (August 3-9 2013) p 21 Table 2

9.  Helm D (2012)  The Carbon Crunch  Yale University Press  pp 195-6